Skip to main content

Full text of "Enhancing productive physical therapy in children using strategic interaction training"

See other formats


ENHANCING PRODUCTIVE PHYSICAL THERAPY IN CHILDREN 
USING STRATEGIC INTERACTION TRAINING 



BY 

STACEYJ. HOFFMAN 



A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL 

OF THE UNrVERSITY OF FLORIDA IN PARTIAL FULFILLMENT 

OF THE REQUIREMENTS FOR THE DEGREE OF 

DOCTOR OF PHILOSOPHY 

UNrVERSITY OF FLORIDA 

1999 



ACKNOWLEDGMENTS 

I would like to express my sincere gratitude to my chairperson, Dr. Stephen R. Boggs 
for his guidance during my graduate school training and his support in completing this 
project. I would also like to thank committee members Drs. Sheila Eyberg, Kathye Light, 
Mary Jane Rapport and Trevor Stokes for their support and encouragement. 

This research could not have been accomplished without the assistance of the 
physical therapists and the children at Kids on the Move, Shriners Children's Hospital and 
All Children's Therapy Center. The research would not have been completed without the 
tremendous efforts and dedication of my research assistants, Stephanie Strobeck and Karen 
Whidock, who provided coundess hours of coding and videotaping. I am also grateful to 
the Center for Pediatric Psychology and Family Studies for providing partial funding for this 
project. 

I am especially thankful for the ongoing support of my family and friends. In 
particular, I am indebted to my husband and friend, John Barbour, for his never-ending 
patience, support, and encouragement; to my best friend Murphy; to Ardis Hanson for her 
friendship, generosity, and assistance in editing this manuscript; and to my parents Sandra S. 
Hoffman and William R. Hoffman for their emphasis on education and faith in my abilities. 



TABLE OF CONTENTS 

gage 

ACKNOWLEDGEMENTS ii 

LIST OF TABLES v 

LIST OF FIGURES vi 

ABSTRACT vii 

INTRODUCTION AND REVIEW OF THE LITERATURE 1 

Behavior Modification for Motor Skills 5 

Behavior Problems in PT 12 

Training Programs in Behavior Modification 14 

SPECIFIC AIMS AND HYPOTHESES 21 

Therapist Target Behaviors 21 

Child Target Behaviors 22 

Consumer Satisfaction Predictions 22 

METHOD 23 

Participants 23 

Setting 25 

Measures 26 

Procedure Design 34 

RESULTS 38 

Therapist Behaviors 38 

Child Behaviors 47 

Consumer Satisfaction 60 

DISCUSSION 62 

Differences between settings 65 

Strengths and weaknesses of the study 67 

Implications 68 

Future directions 69 



page 
APPENDICES 

A PROCEDURAL OUTLINE FOR CONDUCTING SIT TRAINING 74 

B DEMOGRAPHICS QUESTIONNAIRE 81 

C CONSUMER SATISFACTION QUESTIONNAIRE 82 

D BEHAVIOR DEFINITIONS 84 

REFERENCES 89 

BIOGRAPHICAL SKETCH 96 



IV 



LIST OF TABLES 

Tabic page 

1 Child demographics, diagnosis and behavioral issues grouped by treating 
therapist 24 

2 Summary of reliability for behavior categories 33 

3 Mean occurrence of target behaviors at baseline and intervention for all 
therapists 39 

4 Average change in the inpatient setting vs. outpatient setting 59 

5 Average change in the inpatient setting vs. outpatient setting 60 

6 Mean scores on Consumer Satisfaction Questionnaire 61 



LIST OF FIGURES 
Figure page 

1 Percent occurrence of direct commands - inpatient setting 40 

2 Percent occurrence of direct commands - outpatient setdng 41 

3 Percent occurrence of ignoring inappropriate behavior - 

Inpatient setting 42 

4 Percent occurrence of ignoring inappropriate behavior - 

outpatient setting 43 

5 Percent occurrence of all attention to appropriate child behavior - 
inpatient setting 45 

6 Percent occurrence of all attention to appropriate child behavior - 
outpatient setting 46 

Rate of praises - inpatient setting 48 

8 Rate of praises - outpatient setting 49 

9 Percent occurrence of compliance— inpatient setting 51 

10 Percent occurrence of compliance— outpatient setting 52 

1 1 Percent occurrence of noncompliance— inpatient setting 53 

1 2 Percent occurrence of noncompliance— outpatient setting 54 

13 Rate of inappropriate child behavior— inpatient setting 56 

14 Rate of inappropriate child behavior— outpatient setting 57 



Abstract of Dissertation Presented to the Graduate School of the University of Florida in 
Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy 

ENHANCING PRODUCTTVE PHYSICAL THERAPY IN CHILDREN 
USING STRATEGIC INTERACTION THERAPY 

By 

StaceyJ. Hoffman 
December 1999 

Chairperson: Stephen R. Boggs, Ph.D. 

Major Department: Clinical and Health Psychology 

Many children with developmental disabilities or orthopedic impairments require 
physical therapy (PT) to help maximize function, independence, and mobility. Children with 
developmental disabilities may also have impaired cognitive abilities and a higher incidence 
of negative behavior that can interfere with PT. Negative behaviors such as noncompliance, 
resistance, tantrums, and other delay tactics can often prevent the therapist from 
implementing a treatment program and thus impede the process of PT. Yet without PT, 
these children are at increased risk for progression of their disabilities and immobility as well 
as further impairment of their social development. 

The present study used a training intervention to teach physical therapists behavioral 
techniques to deal more effectively with children who exhibit behavior problems in PT. A 
multiple baseline design was used with two therapists and four children in an inpatient 
children's hospital, and two therapists and three children in an outpatient setting. Results 
indicated that physical therapists learned the behavior modification techniques and 
incorporated these skills into their daily PT session. Results also indicated that therapists 



Vll 



increased their levels of direct commands, praise, attention to appropriate child behavior, 
and ignoring inappropriate child behavior. Further, child compliance rates increased, 
and inappropriate child behaviors decreased. The intervention was rated positively 
by all therapists. 



Vlll 



INTRODUCTION AND REVIEW OF THE LITERATURE 

Many children with developmental disabilities or orthopedic impairments require 
physical therapy (PT). Goals of PT may be to increase or maintain strength or range of 
motion, prevent contractures or other deformities, provide appropriate seating to maximize 
function and promote postural alignment, and improve functional mobility and 
independence. However, many of these children may not receive adequate PT due to refusal 
by private insurance or managed care companies to pay for PT or to authorize adequate 
numbers of sessions. Other children may not receive adequate services due to time 
constraints on the therapist to provide more than standard range of motion exercises. Still 
other children may not receive the PT they need due to behavior problems that interfere 
with the therapy and limit progress towards therapeutic goals. 

Compared with typically developing children, children with developmental 
disabilities often have impaired cognitive abilities, attention deficits, a higher need for 
attention from adults, and a high incidence of negative behavior (Briener & Beck, 1984). 
Conduct problems such as noncompliance, aggression, resistance, crying, tantrums, 
withdrawing, and other delay tactics can often prevent the therapist from implementing a 
treatment program and thus impede the process of PT. Studies using functional analyses to 
identify influences on problem behaviors found that behavior problems occur more 
frequently in high demand situations (Repp & Karsh, 1994; Carr, Newsom, & Binkoff, 
1980). Carr and Durand (1985) found that inappropriate behaviors increased during high 
demand situations and in the absence of adult attention. Physical therapy is a high demand 



situation requiring the patient to perform multiple exercises and activities that are often 
difficult and may be painful. Thus, the pediatric physical therapist is frequendy faced with 
children who are aggressive and noncompliant often without the skills to handle such 
behavior. Yet without PT, these children are likely to be at increased risk for progression of 
their disabilities and immobility as well as further impairment of their social development. 

There is a plethora of literature focusing on interventions for aberrant behaviors in 
persons with developmental disabilities (Repp & Karsh, 1994). Such research has typically 
been conducted by psychologists and has been conducted in a variety of settings such as 
schools (e.g., Repp & Karsh 1994), the dental office (Allen & Stokes, 1987), the home (Day, 
Horner, & O'Neill, 1994), inpatient settings (Mace, Page, Ivancic, & O'Brien, 1986), and 
during PT sessions (Rapport & Bailey, 1985). 

According to the literature (Carr, 1977; Durand & Carr, 1985), four primary 
motivating conditions for engaging in aberrant behaviors such as self-injurious and 
aggressive behavior have been identified. These include attention seeking behaviors, escape 
motivated behavior, sensory motivated behaviors (e.g. self-injurious behavior; SIB), and 
obtaining tangible consequences such as food or a toy. Escape and attention have been 
found to be the most common motivating conditions (Iwata et al., 1994). 

The appearance of aggression and noncompliant behavior in PT is not uncommon. 
Reynell (1 965) found an interfering, negative behavioral response pattern in postoperative 
children with cerebral palsy. One of the more notable behaviors reported during treatment 
observations, described by both the therapists and the children's parents, was aggression 
toward the therapist. Lalli, Mauk, Goh, and Merlino (1994) and Manella and Varni (1981) 
noted that problem behaviors such as aggression and self-injurious behavior increased when 
the patients were requested to participate in their PT sessions. Horner (1971) reported the 



occurrence of aggression in a 5 year old child with spina bifida and mental retardation as the 
requirements for walking were increased. Carr, Newsom, and Binkoff (1980) found that 
aggression in children with retardation increased in demand situations. Furthermore, it was 
shown that aggression was maintained by negative reinforcement in the form of escape from 
demands. The general conclusion has been that aggression often functions as an escape 
response in such high demand situations and when successful, is maintained through 
negative reinforcement. Depending on the physical therapist's response to such behaviors, 
the child may adapt and progress physically or the child may regress in his/her behavioral, 
emotional or physical repertoire (Stokes, Mowrey, Dean, & Hoffman, 1997). 

Not only might physical therapists have to overcome overt behaviors such as 
aggression, tantrum behavior, resistance, and pain behaviors, but they might also deal with 
other delay tactics that impede the flow of therapy. Such problems may include inattention, 
self-stimulating behaviors, noncompliance, and dawdling such as stopping an exercise to ask 
questions or to engage the therapist or others in conversation. These behaviors are equally 
interfering with the process of PT and increase the amount of time spent in therapy while 
decreasing productivity. 

Behavior therapy has been shown to be an effective approach to evaluating and 
changing behavior. Behavior modification techniques have long been used with children and 
adults with cerebral palsy, mental retardation, autism, and other developmental disabilities to 
treat behaviors such as self-injurious behavior (Durand & Carr, 1985), aggression (Carr, 
Newsom, & Binkoff, 1980), noncompliance (Rapport & Bailey, 1985), pain behaviors 
(Fordyce, Shelton, & Dundore, 1982), and others. However, the use of behavior 
modification techniques in PT by physical therapists has been slow to evolve. 



4 

The use of behavior modification in PT can serve to modify the environment to 
reduce difficult behaviors that interrupt therapy. Yet, the majority of applications of 
behavior modification in the PT setting reported in the literature have focused on using 
behavior modification to increase motor skills (Martin, 1976). Additionally, such 
interventions appear to have been conducted by psychologists and not the physical 
therapists. However, the majority of children with physical disabilities are treated by physical 
therapists, not behavior therapists. Few attempts have been made to train physical therapists 
in behavioral techniques to modify children's behavior. 

Lalli, Mauk, Goh, and Merlino (1994) successfully trained two physical therapists to 
use a behavioral intervention to increase compliance to commands, decrease rates of 
problem behaviors, and increase the distance ambulated in PT. The training of the physical 
therapists occurred after the behavior therapist had implemented the intervention and had 
determined its effectiveness. Unfortunately, the authors failed to describe how the therapists 
were trained and over what period of time, nor was there any follow-up to determine if the 
therapist's training carried over to children exhibiting aberrant behaviors who were not 
participants in the study. 

By incorporating behavioral techniques into the physical therapist's repertoire, the 
therapist may be able to decrease negative behaviors that interfere with PT. Additionally, 
behavioral techniques such as positive reinforcement could result in higher rates of success 
and lead to greater treatment gains in fewer sessions. 

The remainder of the chapter critically examines the literature regarding the use of 
behavior modification to promote the development of motor skills and to improve behavior 
and participation in PT. Early studies in the area (Fuller, 1949; Harris, Johnston, Kelley, & 
Wolf, 1964; Rice, McDaniel, & Denney, 1962; and Trotter & Inman, 1968) focused on using 



behavior modification to promote changes in motor skills. It was generally concluded that 
such use of behavior modification in the treatment of persons with physical and mental 
disabilities was effective. As Manella and Varni (1981) pointed out, however, the use of 
behavior modification in physical rehabilitation can be used not only to promote optimal 
physical functioning, but also to promote modification of behaviors that often occur in 
children with long-term disabilities and that may interfere with FT. Literature on training 
behavior change agents is reviewed as well. The review concludes with a summary and 
critique of existing literature followed by a discussion of the hypotheses to be examined 
in this research. 

Behavior Modification for Motor Skills 

Perhaps the earliest reference to the use of behavior modification to increase 
movement in a person with a disability was made by Fuller (1949). Fuller described the use 
of operant conditioning with an 1 8-year-old male with profound impairment. By using a 
squirt of sugar-milk solution each time the subject moved his arm, Fuller was able to 
increase the rate of upper extremity movements from .67 per minute to 3 per minute. 

Almost twenty years later, Foss (1966) described several studies that provided 
evidence for the successful use of behavioral techniques to control movement in humans, 
and suggested its use for children with cerebral palsy. Early literature addressing behavior 
modification with children who have cerebral palsy (CP) continued to focus primarily on the 
promotion of learning motor skills. Rice, McDaniel, and Denney (1962) described a single- 
case study in which reinforcement of arm movements was provided to a six-year-old boy 
with profound mental retardation. After they increased the operant rate of arm movements 
by rewarding the subject for any arm movements, the authors used ice cream as a reinforcer 



to shape the response of reaching for a ring placed over the subject's bed. With ice cream as 
a reinforcer, the subject touched the ring over 700 times during a 45-minute period. 

Early reports of the use of behavioral techniques in the PT environment were 
provided by Meyerson, Kerr, and Michael (1967). These authors presented four case studies 
demonstrating the use of behavior modification in a rehabilitation setting. The target 
behaviors included developing independent ambulation with two subjects, promoting on- 
task behavior during a typing task with one subject, and decreasing high rates of self- 
injurious and self-stimulating behavior with another subject. In each case, a behavior 
therapist designed and implemented specific interventions for each subject. Interventions 
included shaping target behaviors through successive approximation, providing positive 
reinforcement of target behaviors with attention from adults or edible rewards, and 
providing alternative sensory stimulation for non-injurious behavior. Each intervention was 
successful in achieving the target goal. However, the physical therapists and the 
occupational therapist who had been treating the children prior to the research intervention, 
were not incorporated into the study to learn the interventions themselves. 

Trotter and Inman (1968) described the use of positive reinforcement to promote 
the performance of progressive resistance exercises in patients with paraplegia and 
quadriplegia involved in physical rehabilitation. By providing positive reinforcement in the 
form of attention from the therapist, and increased attention to appropriate rehabilitation 
goals by charting, the authors hypothesized that the group receiving positive reinforcement 
would achieve greater gains in their strengthening program. Indeed, a significant difference 
was found in the average increase in weights lifted for the experimental group versus the 
control group. However, due to several limitations of the study, the reasons for these 
differences could not be isolated. Differences between the control and experimental group 



other than the use of positive reinforcement significantly limited the conclusions that could 
be drawn from this study. For example, half of the control group (subjects with 
quadriplegia) performed their exercises in the presence of and with the assistance of a 
nonprofessional who also provided attention in the form of "usual conversation." The 
experimental group received attention from a trained physical therapist. Status, knowledge 
and experience of the physical therapist versus the nonprofessional aide could contribute to 
the differences observed. Additionally, the control group not only failed to receive positive 
reinforcement, but also failed to receive any prompting or advice to increase the weights 
being lifted. Such differences make it difficult to delineate which component of the 
treatment was most instrumental in creating change. 

Trotter and Inman's (1968) assertion regarding the benefits of using behavior 
therapy techniques in the PT setting seemingly did little to promote the practice. In fact, 
when one considers the tremendous success in using behavior modification to effect motor 
skills as well as behavior in general, it is quite unfortunate that the PT literature remains 
limited in its presentation of the successful combination of behavior therapy and PT. 

In the PT literature, Kolderie (1971) provided a brief literature review related to 
behavior therapy techniques and then described how behavior modification could be used 
with children with cerebral palsy to learn motor skills. A single-case study was described 
involving a seven-year-old, nonverbal female of average intelligence with mild spastic 
diplegia. At the time of the study, the child had a vocabulary of 10 words, could attain a 
sitting position with assistance, was learning to crawl, and could ambulate with maximum 
assistance. Using a poker chip reward system and immediate verbal reinforcement in 
physical, occupational, and speech therapy, the subject progressed to independent 
ambulation and established an age appropriate vocabulary. Over the course of treatment, 



8 

reinforcements were gradually shifted to natural contingencies. However, because there was 
no baseline or reversal of condngencies, the authors could not be conclusive in stating that 
behavior modification was the critical variable accounting for the changes in the child's 
behavior. Additionally, no follow-up assessment was done to determine if the progress 
continued. Furthermore, the author did not describe if the physical, speech and 
occupational therapists were trained to implement the intervention or if a behavior therapist 
conducted the treatment. 

In a single-case design, Chandler and Adams (1972) used a behavior modification 
program to promote independent ambulation in an eight-year-old boy with multiple 
handicaps, who would not ambulate despite being physically capable of doing so. During 
the baseline period, the highest mean number of steps obtained from nine trials became the 
base rate. Implementation of the modification phase involved providing the child a reward 
of music or chocolate, contingent on exceeding the base rate from each previous session. 
Independent ambulation was achieved after twenty-eight sessions, and continued after the 
reinforcers were withdrawn. Several problems were evident in this study. Primarily, the 
target behavior being studied, number of steps, varied tremendously across time. 
Additionally, the target behavior had not achieved a stable rate during the baseline phase and 
before the intervention was begun. During the intervention, the number of steps could 
range between 37 and 85 over a span of about five days. This inconsistent performance led 
the investigators to change the reinforcement system three times. Another shortcoming of 
the study was an inadequately described baseline condition. It was not clear if the baseline 
condition was a true baseline based on a typical PT session, or if the setting, or the physical 
therapists' behavior (verbal encouragement or physical assistance) were different. 



Bragg, Houser, and Schumaker (1975) examined the effects of rewarding appropriate 
sitting posidons in children with cerebral palsy in a multiple baseline design across six 
children. Baseline sessions consisted of engaging the child in structured play during which 
no comments were made regarding the way the child was sitdng. The percentage of 
appropriate and inappropriate sitdng and the length of dme spent in each position were 
recorded. The subjects were then randomly assigned to one group that received contingent 
praise, affection and food, or a second group that received consistent priming to sit with 
their legs in front of them in addition to the praise, affection, and food. Appropriate sitting 
was increased in all children and it was also believed to generalize to the classroom setting. 
However the authors failed to obtain a baseline rate of inappropriate sitting in the classroom 
prior to the study. Additionally, inappropriate sitting was not completely eliminated for any 
subject, and there was no follow-up phase to assess continued effects. The authors 
hypothesized that because the investigator played with the child regardless of how they were 
sitting they may have inadvertendy reinforced inappropriate sitting positions. 

Pierce and Garland (1 977) used an extensive combination multiple baseline and 
reversal design to compare the effects of four interventions on promoting motor skills in six 
people with physical and mental disabilities. The four conditions were as follows: (1) prior 
instruction only before each motor task, (2) instruction plus social reinforcement when 
performing well, (3) instruction, social reinforcement, and material reinforcement, and (4) 
instruction, social reinforcement, and goal setting. Although each condition was 
differentially effective, social reinforcement and material reinforcement were found to be the 
most effective combination for increasing performance levels. This led the authors to 
conclude that the material reinforcement was the critical component in that intervention. 
Despite the proven effectiveness of the intervention, all subjects performance levels returned 



10 

to baseline once a procedure was removed. It was recommended that future studies assess 
how to maintain motor skills after reinforcement is removed. 

According to Hester (1981) ambulatory deficiencies in children with profound 
mental retardation may be the result of misplaced reinforcement contingencies. He 
described how nonadaptive behaviors may be inadvertently reinforced, whereas, adaptive 
behaviors may not be reinforced at all. If indeed deficient skills had been learned through 
accidental conditioning, then Hester hypothesized that behavior modification using physical 
guidance and positive reinforcement could significantly affect change in such behaviors. 

In a multiple-baseline design, Hester (1981) described how he used positive 
reinforcement consisting of physical, social, and edible rewards to increase the standing and 
walking behaviors of an institutionalized, 14 year old female, with profound mental 
retardation. During the course of training, as the child made progress, the physical and 
edible rewards were slowly withdrawn. A strength of this study was the addition of the 
generalization phase following the intervention, during which another staff member 
observed the procedure once before duplicating it for two 30-minute sessions each day. 
During the generalization phase, the behaviors initially decreased, but returned to 
appropriate levels. An additional strength in this study was the apparent use of the physical 
therapist in implementing the training. Unfortunately, despite the therapists attempt to train 
other staff members, the high staff turnover rate resulted in inconsistent use of the 
procedure. Thus, at two months follow-up, the child's behavior had returned to baseline. 

Hill (1985) also presented a rationale for combining behavior modification 
techniques with PT in the habilitation of patients with cerebral palsy. Similar to Hester 
(1981), she described how a person with cerebral palsy can develop maladaptive motor 
function as a result of learning and interactions with the environment. More specifically, an 



11 

individual with cerebral palsy is likely to be limited motorically and thus lack opportunities to 
practice appropriate movements. Without an adequate foundation of basic motor skills, 
more complex motor skills may never be developed. In addition, because an individual with 
cerebral palsy may experience limited range of motion, weakness, and increased spasticity 
that may also cause increased pain, certain movements may actually be punishing. Thus, 
such an individual will adapt to these problems by avoiding those movements and 
developing abnormal movement patterns that are easier and likely to be reinforced with 
success and accommodate his/her abilities. Such behaviors are often maladaptive, and may 
be resistant to treatment. 

Rapport and Bailey (1985) used a multiple-baseline design across outcome measures 
designed to assess fine and gross motor skills after implementing a home-based behavior 
therapy program combined with an existing PT program. The goals of the program were to 
increase compliance and improve fine and gross motor skills for an 8 year old with spina 
bifida. A unique addition to this study was the involvement of the parents, who were trained 
to use a variety of game-like tasks to practice the skills. In addition, a motivational chart was 
used to indicate those activities that were to be worked on each day and to display a point 
system for earning rewards for each task. Points could be exchanged for a variety of 
reinforcers at the end of each week. Although the authors concluded that the combined 
behavioral intervention with the existing PT program was successful, it was not possible to 
determine the factor(s) primarily responsible for the change. The improvement in this study 
compared to other studies reviewed was the use of a one -month follow-up period that 
indicated the performance gains had increased or remained stable. 

There are several shortcomings in the existing literature on behavior modification 
and PT. First, the studies described only used behavior modification techniques to promote 



12 

the learning of motor skills. What has been lacking in the research conducted with children 
with developmental disabilities who receive PT, is the use of behavior modification to 
address behavioral problems that interfere with PT. A second shortcoming in several of the 
studies reviewed was the inability to conclude that the intervention provided was the primary 
factor affecting change. Third, several studies failed to provide follow-up data to verify that 
such behavioral techniques can provide a lasting effect. Finally there is a lack of attention to 
providing training to other people, such as parents and therapists, who are involved in the 
care and treatment of the children with physical disabilities. 

Behavior Problems in PT 

Perhaps it is because physical therapists focus on the development of motor skills 
and maximizing mobility that most studies have focused on the application of behavior 
modification to the learning of motor skills. However, another aspect of PT in which 
behavior modification techniques can be useful is in dealing with behavior problems. 
Behavior problems such as aggression, noncompliance, and tantrum behaviors, are 
frequently found in PT (Lalli, Mauk, Goh, & Merlino, 1994; Homer, 1971) and often 
interfere with progress in PT. 

Horner (1971) reported the occurrence of aggression in a 5-year-old child with spina 
bifida and mental retardation, as the requirements for walking were increased. The subject 
had received surgical procedures to correct bilateral hip dislocations at the age of four. 
Following his surgery, he received PT on a regular basis that focused on strengthening 
exercises in preparation for walking. Despite ongoing PT, and having the physical 
capabilities to ambulate, attempts at enforced ambulation were met with increased negative 
behaviors such as tantrums and resistance. 



13 

Initially, Horner used a reversal design with a six-step successive approximation 
sequence to establish the use of parallel bars to assist with walking. Root beer had proven to 
be a reinforcer during the baseline phase. When the child was ambulating independently in 
the parallel bars, an extinction procedure was initiated demonstrating control of the 
reinforcement contingency over the behavior. Following the extinction period, the subject 
received a "free" reinforcer and returned to the final step in the initial treatment period. 

After ambulation was obtained in the parallel bars, the use of crutches was 
established using a ten-step successive approximation sequence. During the baseline phase 
for this stage of the treatment, aggressive behaviors started to occur, such as throwing 
crutches and tantrum behavior. The subject was prevented from throwing the crutches by 
securing them to his hands with elastic bandages, and a 3-minute time out was enforced 
contingent upon his aggressive behavior. Although systematic records of the subject's 
resistive behavior were not kept, the author stated that these behaviors responded to the 
time-out and use of restraint. Also during this phase of treatment, the reinforcer was 
changed after session 54 from root beer to noise that the subject was allowed to make after a 
trial was successfully completed. This was reportedly due to the difficulty in being able to 
immediately reinforce correct responses. Independent ambulation was achieved, after which 
a contingency management program was designed to ensure continued independent 
ambulation. The sound methodology allows one to conclude that behavior modification 
was indeed the critical variable in training this child to ambulate independently in the 
parallel bars. 

In a single-case study, Manella and Varni (1981), described a four and one-half-year- 
old girl with myelomeningocele who presented with a short attention span, tantrum 
behaviors and refusal to walk during PT. Using social attention, affection, and game playing, 



14 

positive reinforcement was contingent upon the partial performance of certain motor skills. 
As the child became more skilled, the amount of assistance needed was decreased, and 
independence was reinforced. All inappropriate behaviors were ignored. After the child 
exhibited appropriate behaviors in PT, the child's mother was trained in the behavioral 
techniques for home use. Results indicated that the child reached the therapeutic goals set 
by the physical therapist. Additionally, the mother was effective in implementing the 
behavioral program in the home resulting in continued progress. As might be expected, the 
child's ambulation deteriorated when the mother began implementing the program, but 
quickly returned to and then exceeded the performance she had obtained with the physical 
therapist. A limitation of this study was that the frequency of the problem behaviors that 
had previously been exhibited before the onset of the study were not mentioned after the 
initial description. The occurrence of the behaviors was not documented in the study other 
than to state that they were ignored. Thus, it is not known if these behaviors were 
eliminated before, during, or after the intervention, or were not eliminated at all. However, a 
strength of the study was training the parent in the application of the behavioral technique. 
Such training was an essential component in the treatment, and one that has rarely been 
addressed in the literature. 

Training Programs in Behavior Modification 
Programs that train parents and healthcare professionals in behavioral procedures for 
children have been shown to promote more positive parent-child interactions at home and 
optimize health care provision in the hospital (Eyberg, Boggs, & Algina, 1995; Babbitt et al., 
1994; Singer, Nofer, Benson-Szekely, & Brooks, 1991). Parent training has probably 
received the most attention in the literature. Various child-treatment procedures have been 
taught in parent-training programs. Most frequently, parents have been trained to apply 



15 

differential reinforcement either as the only form of treatment or with other procedures 
(Budd, Green & Baer, 1976). One of the key advantages of using a behavioral approach to 
deal with the behavior problems of children is the relative ease with which an individual who 
may have relatively little skill in the use of sophisticated therapy techniques can learn and 
implement the basic principles of behavior change (Riley, Parrish, & Cataldo, 1989). 

One of the most common parent training approaches involves the use of modeling, 
practice and immediate feedback (e.g., Forehand and King, 1977). Parent-child interaction 
therapy (PCIT; Eyberg & Matarazzo, 1980; Eyberg, Boggs, & Algina, 1995) for example, was 
developed to treat children with conduct problem behaviors and their families. In PCIT, the 
therapist models for the parents certain skills to be learned. The parents then practice the 
skills with their child, while receiving feedback through a bug-in-the-ear device. Following 
each session, the therapist provides feedback to the parent regarding their performance. 
PCIT outcome research has demonstrated statistically and clinically significant 
improvements in child problem behavior (Eisenstadt, Eyber, McNeil, Newcomb, & 
Funderburk, 1993; Eyberg, Boggs, & Algina, 1995; Eyberg, Robinson, 1982). 

Hudson (1982) conducted a component analysis of a group training program for 
parents of children with developmental disabilities to teach their children new skills. Forty 
mothers were randomly assigned to one of four treatment groups: verbal instruction, verbal 
instruction plus the teaching of behavioral principles, verbal instruction plus the use of 
modeling and role-playing with immediate feedback, and a wait-list control group. Results 
indicated that the inclusion of modeling, role-play, and feedback led to significantly greater 
improvement in the ability of parents to develop new skills in teaching their child. 

Eyberg and Matarazzo (1980) compared two types of training to promote behavior 
management skills in mothers. The didactic group received five, 90-minute sessions where 



!6 

they were taught the principles and application of behavioral techniques. Mothers in the 
individual interaction group were trained in the basic rules of two components of parent- 
child interaction. The rules were taught to the mothers in five, 20-minute sessions, using 
description, modeling, interaction and feedback. Results indicated that mothers in the 
interaction group improved significantly on all targeted behaviors. 

"Train the trainer" techniques are perhaps an inevitable extension of the research 
demonstrating the effectiveness of behavioral approaches to change human behavior. 
Trainers, or behavior change agents, that have been studied include teachers to increase 
social interactions (Hendrickson, Gardner, Kaiser, & Riley, 1993), institutional attendants to 
apply behavioral techniques (Gardner, 1972), graduate students to train parents (Isaacs, 
Embry, & Baer, 1982), medical residents to increase communication skills (Branch, 1990), 
and foster grandparents working with institutionalized children (Fabry & Reid, 1978). 

Research on training the trainer has frequently focused on the most effective ways to 
train others in the implementation of behavioral techniques (Bernstein, 1982). In a review of 
consumer satisfaction in parent training programs, practice with the child was found to be 
the most useful training tool, followed by performance-oriented teaching, and trainer 
demonstration. The use of written materials was found to be least useful (McMahon & 
Forehand, 1983). Gardner (1972) found role playing to be more effective than lecture in 
teaching behavior modification techniques to nonprofessionals. Cunningham, Davis, 
Bremner, Dunn, and Rzasa (1993) found modeling, role play, and homework most useful in 
promoting mastery of new skills. Hosford and Johnson (1983) found that using the self-as- 
model technique, with only appropriate behavior viewed, resulted in completely 
extinguishing inappropriate interviewing behaviors in counselors. 



17 

It is generally agreed upon that instruction alone is insufficient to teach behavior 
change skills (Bernstein, 1982; Delamater, Conners, & Wells, 1984; McMahon & Forehand, 
1983). However, sufficient support exists for modeling (Bandura, 1969; Cunningham et al., 
1993), rehearsal, role play, and feedback (Delamater, Conners, & Wells, 1984). 

Modeling has repeatedly been shown to be a powerful technique for behavior 
change. Effective modeling has been used to teach speech and language skills (e.g. Charlop 
& Milstein, 1989), improve interviewing skills (e.g. Miltenberger & Veltum, 1988), acquire 
motor skills (e.g., Carroll & Bandura, 1985), reduce avoidant behavior (e.g., Meichenbaum, 
1971), teach parenting skills (e.g., Webster-Stratton, 1981), and reduce child uncooperative 
behavior (e.g., Stokes & Kennedy, 1980). Symbolic modeling in the form of videotapes was 
the logical outgrowth of the documented effects of live modeling. Videotape modeling has 
several advantages over live modeling, including the ability to create naturalistic modeling 
sequences, greater control over creating the modeling scene, ability to present multiple 
models or repeated observations of the same model, efficiency, and self-administered 
treatment sessions (Thelen, Fry, Fehrenbach, & Frautschi, 1979). 

For a videotape model to be effective, the modeling procedure should focus on the 
skill to be learned, its context, and its consequences. Bandura (1969) identified four 
components that mediate observational learning: attention to modeled events, retention of 
what is observed, ability to replicate modeled behaviors, and motivation to reproduce those 
behaviors. The characteristics of the model contribute to the effectiveness of the procedure. 
Model characteristics such as competence (Baron, 1970), status level (McCullagh, 1986), age 
(Bandura & Kupers, 1964), and similarity (Kazdin, 1974) have been shown to strengthen the 
degree of observational learning. 



18 

The empirical evidence for the use of videotaped modeling allows it to stand alone as 
an effective intervention in its own right. However, packaging it with other effective 
interventions, especially with opportunities to practice, should be even more effective 
(Dowrick, 1991). Bandura (1971) suggested that the effects of modeling could have a greater 
impact if it were followed with guided practice or rehearsal in a natural environment. 

In his review of the parent training literature, O'Dell (1985) argued that the best 
parent training programs have been a combination of visual, verbal and interactive training 
methods. Kazdin (1994) also promoted the use of multicomponent interventions for several 
reasons: (1) combining individual techniques that have been shown to be effective should 
strengthen their impact; (2) a combination of techniques may increase the rate of skill or 
knowledge acquisition; and (3) a multicomponent intervention may be able to address a 
range of problems that often occur together. 

Forehand and King (1977) used cueing, feedback, instruction, role play, and practice 
with feedback to train parents of noncompliant children. After observing the parent-child 
interaction, the parent was instructed on ways to change his/her behavior to increase the 
child's compliance, and the therapist modeled these behaviors. The parent was then able to 
practice the skills with the therapist who provided feedback. Finally, the parent used the 
skills in playing with his/her child while receiving further feedback from the therapist via a 
bug-in-fhe-ear device. Results indicated that the behavioral criterion for the study were 
achieved in an average of nine sessions. Additionally, follow-up assessments indicated that 
the behavior changes were maintained at three months. 

Matson and Stephens (1978), and Delamater, Conners, and Wells (1984) used 
training interventions that included modeling, behavioral rehearsal and feedback to train 
staff to use behavioral techniques in their interactions with children on an inpatient 



19 

psychiatric unit. Webster-Stratton (1990) found greater effectiveness in decreasing deviance 
in children through the use of videotaped modeling plus therapist consultation for parents. 

This chapter has described the literature on the use of behavior modification in the 
field of PT, primarily with children. Behavior modification has been used most frequently to 
facilitate the learning of new motor skills. Although few in number, research studies have 
been conducted demonstrating the use of behavior modification to address behavior 
problems of children during PT. However, the behavior change agents in such studies were 
most often a behavior specialist and the physical therapists were rarely trained in the 
behavior modification techniques. This type of approach continues to leave the physical 
therapist without the skills necessary to deal effectively with the behavior problems that 
frequendy occur in the PT setting. 

A review of the literature on effective components for training nonprofessionals in 
behavior modification techniques was also described. It has been well established that 
parents, teachers, psychiatric technicians, and therapists can successfully learn and 
implement behavioral techniques (O'Dell, 1974; Eyberg & Matarazzo, 1980; Harris, 
Johnston, Kelley, & Wolf, 1964; Rapport & Bailey, 1985; Gardner, 1972). Successful 
training components include modeling (O'Dell, 1974), videotaped modeling with behavior 
therapist feedback (Webster-Stratton, 1990), practice and immediate feedback 
(Eyberg & Matarazzo. 1980). 

Based on the repeated success of training behavior change agents through the use of 
modeling, rehearsal and feedback, it seems likely that a training intervention combining these 
practices would be an ideal training package to promote the acquisition of a new skill. The 
present study used a training module, Strategic Interaction Training (SIT), that incorporates 
videotaped modeling, individual interaction training, and feedback and reinforcement to 



20 



determine if physical therapists can learn to use behavior analysis and therapy reduce 
behavior problems in children during PT. SIT was modeled in part, after Parent-Child 
Interaction Therapy (Eisenstadt, Eyber, McNeil, Newcomb, & Funderburk, 1993; Eyberg, 
Boggs, & Algina, 1995; Eyberg, Robinson, 1982). To date, no other work has trained 
physical therapists to apply general behavioral techniques to the problem behaviors that 
occur in PT. 

Compared to previous studies involving the use of behavior therapy in PT, the 
current study differs in several ways. First, in this study, the physical therapists were 
directly trained to use behavioral modificadon techniques to effect change in the behavior 
problems that children present with during PT. Second, multiple PTs were trained in these 
skills, and the outcome is assessed across several children. Finally, two very different 
settings, an outpatient and inpatient setting, were used to demonstrate the generalizability 
of such training. 



SPECIFIC AIMS AND HYPOTHESES 

This study specifically focuses on training physical therapists in behavior analysis and 
therapy to decrease resistant, noncompliant and aggressive behaviors in children during PT 
and to increase child compliance rates. The training intervention, Strategic Interaction 
Training (SIT), was the independent variable of interest. Dependent variables were the 
frequency of target behaviors coded for both the physical therapist and the child across 
sessions, and social validity measures. It was believed that the results of this study would: (1) 
identify problem behaviors that physical therapists frequendy encounter; (2) show that 
physical therapists can learn to use behavior analysis and therapy; (3) demonstrate that the 
use of behavioral techniques by the physical therapist will result in a decrease of negative 
behaviors in the children receiving PT. 

Specifically, the following hypotheses were tested regarding the 
aforementioned variables: 

Therapists Target Behaviors 

(1) Prior to SIT, it was expected that physical therapists would exhibit low 
frequencies of the following behaviors: praise, direct command, ignore, attention to 
appropriate behavior. 

(2) It was hypothesized that physical therapists could learn to implement behavioral 
techniques to address problem child behavior. Specifically, it was expected that the physical 
therapist would increase rates of the following behaviors to a pre-determined goal level: 
praise, attention to appropriate child behaviors, ignore, and direct commands. 



21 



22 



(3) Additionally, in follow-up sessions, it was hypothesized that the training would 
promote the use of the skills with other children who present with negative behaviors in PT. 

Child Target Behaviors 
As a result of the physical therapists training, the following hypotheses were made 
regarding the behavior of children being treated. 

(4) Child compliance rates would increase. 

(5) Child inappropriate behaviors (e.g., whining, crying, aggression, pain behaviors, 
yelling, and smart talk) would decrease. 

Consumer Satisfaction Predictions 

(6) On a measure of consumer satisfaction, it was predicted that the physical 
therapists receiving SIT would rate it as an acceptable, effective, and beneficial training 
module to reduce behavior problems in children and increase compliance during PT. 



METHOD 
Participants 

Four currently practicing, licensed pediatric physical therapists provided informed 
consent and participated in the training. There were two therapists from an outpatient 
setting and two therapists from an inpatient setting. All physical therapists were Caucasian 
females and had received a bachelors degree in PT. The average number of years practicing 
PT was 15 years and 3 months (range 6-20 years). The average number of years practicing 
PT with children was 11.5 years (range = 6-18 years). One therapist from each site was 
certified in neurodevelopmental therapy (NDT). One therapist at Shriners treated three 
children in the study, one therapist at All Children's Therapy Center treated two children 
while the other two therapists each treated one child at each site. 

Table 1 presents information on the children and the grouping of children with each 
therapist. A total of 7 children were recruited for the study, 4 from Shriners Hospital and 3 
from All Children's Therapy Center. The parent(s) of each child provided informed consent 
to participate, and for those children over 7 year of age, assent was obtained from the child. 
Six children were Caucasian, one child, Sherry, was African American. There were 3 males 
and 4 females, and they ranged in ages from 3 years to 12 years old (M = 7.9). Each child 
was nominated by his/her physical therapist, who was also participating in the study, based 
on the presence of behavior problems that interfered with the process of PT. Children in 
the inpatient setting had been working with their therapist for 1 -2 months prior to entering 
the study. Children in the outpatient setting had been working with their therapists for 



23 



24 



Table 1 



Child demographics, diagnosis and behavioral issues grouped by treating therapist . 



Therapist Participants Gender Age 



Medical 
Diagnosis 



Behavioral Issues 



Number of 
Post-SIT 



1 Sierra 

Shriners 
Hospital 



Sherry 



2 

Shriners 
Hospital 



All 

Children's 

Therapy 

Center 



ah 

Children's 

Therapy 

Center 



Ron 



M 



Alan 



M 



Katy 



8 multiple congenital 

anomalies including 
tibia hemiamelia with 
resultant leg length 
discrepancy on left, 
treated with an 
orthofix 

8 spastic diplegic 

cerebral palsy; status 
post bilateral 
hamstring lengthening 

12 Legg-Calve Perthes 
disease of the left leg; 
status post left 
adductor release and 
varus derotational 
osteotomy 
approximately 5 
months prior to his 
admission; ADHD 

3 spastic quadriplegic 

cerebral palsy; status 
post bilateral adductor 
releases and bilateral 
tendo achilles 
lengthening 

8 spastic diplegic 

cerebral palsy and a 
seizure disorder 



Bryan 



M 



Man 



spastic diplegic 
cerebral palsy; status 
post bilateral 
hamstring lengthening 



Downs syndrome, 
juvenile rheumatoid 
arthritis, and 
neurofibromatosis. 



whining, 
noncompliance, 
lack of 
independence with 



screaming, crying, 

noncompliance, 

aggression 

noncompliance, 
impulsivity that 
was disruptive to 
the flow of PT, 
and whining 



10 



refusing to walk, 
screaming, 
tantrums; easily 
distractible 



suddenly 
becoming very 
limp and falling 
over or collapsing, 
whining. 
noncompliance, 
talking back to the 
therapist and 
crying 

whining, talking 
back to the 
therapist, 
aggression, and 
general 
noncompliance 

noncompliance, 
and talking back to 
the therapist 



25 

approximately two years prior to their involvement in the study. Child behavior problems 
included noncompliance, whining, screaming, aggression, becoming limp and falling over or 
falling to the floor, talking back to the therapist, and dawdling. 

Because of the need for the child to understand language in order to assess 
compliance with therapist requests, children with a history of moderate to severe mental 
retardation were not included and the ability to understand therapists requests was 
determined by observing the child/therapist interactions prebaseline. If any therapists 
achieved training criterion during the first two baseline sessions on the target behaviors to be 
described they were excluded from the study. No therapist met all criterion during baseline. 

Setting 

Participants were recruited from and treated in two different settings. Two physical 
therapists and four children were recruited from Shriners Hospital for Children in Tampa, 
Florida. Shriners is a 60-bed pediatric orthopedic, inpatient hospital that serves children 
from around the state of Florida, southern Georgia, and the Caribbean. For the purposes of 
this research, PT was conducted in a smaller gym adjacent to the main treatment gym and 
away from the noise and distractions of other people. This was done for several reasons. 
First, by using a quieter room, we were able to obtain a better sound for coding therapist and 
child behaviors. Second, it prevented the first therapist from influencing the second 
therapist's behaviors after completing the intervention. Finally, baseline and treatment were 
both conducted in the smaller gym in order to maintain a stable environment during baseline 
and the intervention. Ambulation activities occurred in the main hall, which was typical for 
that setting. Because Shriners Hospital is an inpatient rehabilitation setting, each child was 



26 

seen for PT every day, excluding weekends. The time at which each child was seen during 
the day was relatively constant throughout the research, and each child was seen an average 
of 60 minutes, 5 days a week. 

Two therapists and three children were recruited from All Children's Therapy 
Center, an outpatient therapy center affiliated with a large children's hospital located in St. 
Petersburg, Florida. Physical therapy at All Children's Therapy Center was conducted in 
one large room, often with other children and therapists present. Because of the physical 
layout of the setting, it was not possible to separate the therapists. Two children were 
scheduled for PT once a week, one for thirty minutes and one for one hour, and one child 
received PT three times a week for thirty minutes each. Additionally, one child scheduled 
for PT once a week, had a very high no-show rate resulting in the therapist having less 
consistent opportunities to utilize her skills with this child and implement feedback she 
received on a regular basis. Both PT departments were equipped with similar PT equipment, 
which included bolsters, therapy balls, mat tables, and small staircases. 

Measures 
Observational Measures 

Each PT session was videotaped during all phases and coded for target behaviors to 
be described below. The coding system consisted of portions of the Dyadic Parent-Child 
Interaction Coding System - II (DPICS-II; Eyberg, Bessmer, Newcomb, & Edwards, and 
Robinson, 1994), and revised portions of OTIS (Observation, Training, and Interaction 
System-revised; Stokes & Mowrey, 1999), which was adapted from the Systematic Carousel 
Observation of Performance (SCOOP; Osnes & Stokes, 1987), as well as the addition of a 



27 



pain behavior category that was not covered under DPICS-II or OTIS. Two sequences of 
behavior were also coded: 1) the child's response to the therapist's commands and 2) the 
therapist's response to the child's inappropriate or pain behavior. 

The DPICS-II is a revised and expanded version of the DPICS (Eyberg & Robinson, 
1983), a behavioral coding system designed for the assessment of the quality of parent-child 
social interactions. It was originally designed to be used as a pretreatment and posttreatment 
observational measure, as well as a measure of treatment progress and outcome. DPICS-II 
and its categories were designed for multiple purposes and can be modified depending on 
the needs of the user. Its reliability and validity have been well supported in a recent study 
by Bessmer (1996), who found that the reliability estimates of the DPICS-II codes used in 
this study fell in the good (% agreement >70%; kappa = .60 - .75) to excellent 
(kappa >.75) range. 

The SCOOP (Osnes & Stokes, 1987) was originally designed to be used in diverse 
settings as a measure of the contingencies in effect in various situations. OTIS was adapted 
from SCOOP and was designed for a program serving children with physical impairments. 
There are no existing data on the reliability or validity of this observation and coding system. 
Behavioral Definitions 

The child behaviors that were coded included responses to commands, inappropriate 
behavior, and pain behavior and were defined as follows: 

Compliance (CO; DPICS-II). Compliance with instructions or commands was 
coded when the child obeying or beginning to obey a command within 5 seconds after 
it was given. 



28 

Noncompliance (NC; DPICS-II). Noncompliance is coded when the child does 
not begin to complete an instruction, or demonstrates a behavior that is clearly incompatible 
with compliance of a command within five seconds of the request. 

No opportunity (NOC; DPICS-II). After a therapist issues a command, the child 
has five seconds to respond. No opportunity is coded when the child is not given an 
adequate chance to comply either because the therapist completes the behavior for him/her, 
or the command is clearly out of the range of the child's ability. 

Inappropriate behaviors (IB; OTIS). Inappropriate behavior includes behavior 
that is distracting, off-task, or disruptive including crying, whining, yelling, smart talk, 
aggression. Crying consists of inarticulate utterances of distress (audible weeping) at or 
below the loudness of normal conversation. Whining consists of words uttered by the child 
in a slurring, nasal, high-pitched, falsetto voice. Yelling consists of a loud screech, scream, 
shout or loud crying. The sound must be loud enough so that it is clearly above the intensity 
of normal indoor conversation. Smart talk consists of impudent or disrespectful speech. 
Aggression is any physical touch that is intended to be antagonistic, aversive, hurtful, or 
restrictive of the therapists activities, or harmful to self, or environment. 

Pain behaviors (PB; SIT). Pain behaviors include obvious grimacing, vocalized 
complaints of pain such as yelling out about pain, whimpering, or crying related to pain. 
Inappropriate behaviors and pain behaviors are coded according to the therapist response 
that follows. 

Therapist behaviors that were coded included: 

Direct command (DC; DPICS-II). A direct command is a clearly stated order, 
demand, or direction in declarative form. The statement must be sufficiently specific as to 
indicate the behavior that is expected from the child. 



29 

Indirect command (IC; DPCIS-II). An indirect command is an order, demand, or 
direction for a behavioral response that is implied, nonspecific, or stated in question form. 

Praise (P). Praise includes any specific or nonspecific verbalization that indicates 
liking, approval, or expresses a favorable judgment upon an activity, product, or attribute of 
the child (DPICS-II; combination of labeled and unlabeled praises). 

Ignore (I; OTIS). Ignore is coded when the therapist does not provide any 
evidence of having heard or seen an inappropriate behavior that occurs (passive ignore); or 
the therapist may take action by withdrawing attention, withdrawing physically from the 
child, turning away, or removing an object from the child's reach without responding to the 
inappropriate or pain behavior (active ignore). 

Physical guide (PG; OTIS). If the child is noncompliant following a command, 
the physical therapist physically guides the child to complete the command with minimal 
physical contact. This could include pushing a child's walker forward so that the child 
will continue walking or guiding the child's hand using a hand-over-hand motion to reach 
for an object. 

Inappropriate behavior or pain behavior followed by attention (A- or PBA; 
SIT). Inappropriate behavior or pain behavior followed by attention is coded when a 
therapist talks to the child or provides physical touch, other than is therapeutically necessary, 
when the child is exhibiting inappropriate behavior or pain behavior. This includes 
negotiating (offering a reward such as a desired toy, the presence of a parent, or an activity 
contingent on performance), "threatening" (a specific verbalization indicating a negative 
consequence will follow a behavior) the child, or commenting on or questioning the child's 
inappropriate or pain behavior. 



30 

Inappropriate behavior or pain behavior followed by ignore (I- or PBI; SIT). 

Inappropriate behavior or pain behavior followed by ignore occurs when a therapist 
removes attention by withdrawing from the child, turning away from the child, ceasing 
verbalization with the child, or not providing any evidence of having seen or heard the 
inappropriate or pain behavior exhibited by the child. 

Inappropriate behavior or pain behavior followed by escape (E- or PBE; SIT). 
Inappropriate behavior or pain behavior followed by escape is coded when the therapist 
discontinues an activity, demand situation or therapy for more than 30 seconds due to 
inappropriate child behavior or pain behavior or when an activity is completely terminated 
due to inappropriate or pain behavior. 

Attention following appropriate child behavior (A+; SIT). Talk or physical 
contact with the child occurring before, during, or after appropriate child behavior is coded 
as attention following appropriate child behavior. 
Summary Behaviors 

Two behavior categories, while initially coded separately, were combined in the final 
summary of behaviors. Attention to pain behavior was subsumed within attention to 
inappropriate behaviors. Although it is certainly recognized that pain often occurs during 
physical therapy and can be an important indicator for physical therapists to assess pathology 
or the extent to which a patient can be pushed in his/her exercises, depending on the 
response following a pain behavior, pain behaviors can also serve as a nurturance trap for 
attention and escape (Stokes, Mowery, Dean, & Hoffman, 1997). The criterion level for 
ignoring inappropriate behaviors still allowed for therapists to acknowledge 25% of 
pain behaviors. 



31 

Coder Training 

The primary author served as the primary coder. A second coder who was blind to 
the research hypotheses, served as a reliability coder. Before beginning to code videotapes 
for the current study, each coder completed a minimum of 30 hours of training. Coder 
training consisted of (1) reading and studying relevant portions of the DPICS-II Manual 
(Eyberg, et al., 1994) and the OTIS behavioral definitions. (2) discussion of DPICS-II, 
OTIS, and the other behaviors described; and (3) successfully completing quizzes that 
pertain to each category of the DPICS-II system in The Workbook: A coder training manual 
for the Dyadic Parent-Child Interaction Coding system II (Eyberg, Edwards, Bessmer, & 
Litwins, 1994). Quizzes involving behavior codes not covered in the DPICS II training 
manual were also provided by the primary author. After obtaining 90% on all quizzes of 
relevant behaviors, the reliability coder practiced coding videotapes that had been coded and 
transcribed by the primary coder. Training was considered complete when the reliability 
coder met 80% agreement with the primary coder on two consecutive criterion tapes. The 
reliability coder and the primary coder met on a weekly basis throughout the study to 
practice coding and discuss differences during practice sessions. 
Behavioral Coding 

Each session during baseline, intervention, and follow up was videotaped in its 
entirety and coded for therapist and child behaviors. The outpatient, thirty-minute sessions, 

were blocked into five-minute segments. The 2 n ", 4 tn , and 6" 1 five-minute segments were 
coded for target behaviors. During the inpatient sessions that could be one hour or longer, 
sessions were blocked into ten-minute intervals followed by five-minute intervals throughout 
the session. Frequency coding for target behaviors was conducted during each five-minute 
interval. This allowed for a sample of behavior to be coded from the beginning, middle, and 



32 

end of each PT session. Frequencies were summed for each behavior category and percent 
occurrence or rate per minute was calculated. Percent occurrence was calculated by dividing 
the frequency of a specific category, such as direct commands by the total frequency of the 
behavior category, such as total commands, in each session. The observers and the video 
camera were present for several sessions before initiating data collection to allow the 
participants to adapt to the novelty of a new person and the camera. The PT session and 
timing began when the patient entered the PT gym, and terminated when the patient exited 
the gym at the end of therapy. 
Treatment Integrity 

The primary author conducted the training intervention with each therapist using a 
detailed procedural outline (see Appendix A). Undergraduate research assistants recorded 
treatment integrity data from videotapes of each training session using the outline of the 
intervention in checklist format. Each intervention session was videotaped and coded for 
integrity. Percent agreement was calculated by dividing the number of agreements by the 
number of agreements plus disagreements. Treatment integrity was 100%, indicating that 
the intervention was implemented exactly as it was outlined and indicating a high degree of 
fidelity with the SIT protocol. 
Reliability 

Thirty percent of all intervention sessions were randomly selected to be coded 
independently by an undergraduate research assistant to assess reliability. The 
undergraduate research assistant was naive to the purpose of the investigation. Percent 
agreement was calculated by dividing the number of agreements by the number of 
agreements plus disagreements on occurrences only. Interrater reliability using Cohen's 



33 



kappa (k; Cohen, 1960), which takes into account chance agreement and is thus a more 
stringent reliability measure was also assessed. Reliability was calculated on each behavioral 
category and data are presented in Table 2. 

The mean interrater reliability coefficient was obtained for each behavior category 
and ranged from 77% - 100% and kappa ratings ranged from .82 - .96. Inappropriate or 
pain behavior followed by escape was a very low frequency behavior resulting in an average 
percent agreement of 50% and thus was removed as a dependent variable. 
Table 2 
Summary of reliability for behavior categories 





Behavior category 


% agreement 


Kappa 


Direct Command 


91 


.94 


Indirect Command 


SS 


.93 


Compliance 


85 


91 


Noncompliance 


77 


.86 


No Opportunity for Compliance 


S3 


.89 


Inappropriate Behavior followed by 






Attention 


85 


.91 


Ignore 


83 


.90 


Physical Guide 


8(1 


.87 


Attention to appropriate behavior 


93 


.94 


Praise 


93 


.96 


Warning 


100 


1.00 



Demographics and Medical History 

A demographics and medical history form (see Appendix B) documenting the child's 
age, gender, medical history, and goals for PT was completed by each physical therapist. In 
addition, the therapists provided information regarding their educational background, 



34 

number of years practicing as a physical therapist, number of years working with children, 
and whether they were NDT certified. 
Consumer Satisfaction Questionnaire 

Therapist satisfaction with the treatment intervention was evaluated using the 
Consumer Satisfaction Questionnaire (see Appendix C). This 13-item questionnaire was 
administered to each therapist to evaluate various aspects of the training on a 5-point scale 
(1 = negative, 5 = positive), such as relevance of the training intervention, time 
commitment needed for the intervention, and whether or not the therapist perceived 
behavior changes in the children. The questionnaire also included four open-ended items 
that requested the therapist to provide feedback on the degree to which the intervention was 
acceptable to them, the most helpful and least helpful aspects of the training, along with any 
other feedback for the improvement of Strategic Interaction Training. 

Procedure and Design 

A multiple-baseline design across children and therapists (Baer, Wolf, & Risley, 1968) 
was used in each setting to evaluate the effect of Strategic Interaction Training on therapist 
and child behaviors. A multiple baseline design allows for the visual inspection of the data, 
allowing the reader to quickly see the relationship of one set of data to another set of data. 
The visual analysis of data tends to be more conservative and is thus commonly used to 
assess variables that may impact behavior rather substantially (Parsonson & Baer, 1992). 
Additionally, the multiple baseline design is tailored for unique, applied settings and 
circumstances when only small numbers of participants are available. 

The study was first conducted in an inpatient setting with three children being 
treated by Therapist 1 and one child being treated by Therapist 2. Baseline data were 
gathered on the target behaviors for each Therapist/Child dyad. Because of the multitude of 



35 

target behaviors, it was difficult to achieve a point in the baseline where all target behaviors 
were stable. The intervention occurred with Therapist 1 when a stable baseline occurred for 
the majority of target behaviors and baseline observations continued for Therapist 2. Once 
the behavior of Therapist 1 showed a clear change on the target behaviors, Therapist 2 
received the intervention. This design was replicated in the outpatient setting, where 
Therapist 3 who was treating one child, received the intervention first, followed by Therapist 
4 who was treating two children. To assess generalization of the therapists skills across time 
and with non-targeted children, follow-up data were also collected one month 
postintervention. Thus, this study involved two multiple baseline designs with three phases: 
baseline, intervention, and follow-up on therapist skills. 
Baseline 

During the baseline phase, therapists were instructed to conduct their PT sessions as 
usual. No other instructions or recommendations were provided. Each PT session varied 
between children but was similar for each individual child across sessions, and was based on 
each child's individual PT goals. This most often consisted of requests to perform multiple 
repetitions of strengthening and range of motion exercises, and ambulation activities. Play 
activities were often incorporated into each PT session. 
Intervention 

The Strategic Interaction Training was conducted in three phases: 
Phase one —Didactic training and modeling videotape 

During this phase, the physical therapists viewed a videotape of another female 
physical therapist providing PT to two different children who exhibited a variety of negative 
behaviors such as crying, noncompliance, and resistance. The therapist in the videotape 



36 

modeled the use of the following behavioral techniques for which participating therapists 
would receive training: direct commands, praises, attention contingent upon appropriate 
behaviors, and ignoring inappropriate behaviors. 

The physical therapist and the behavioral consultant, the experimenter, then 
discussed the definitions of target behaviors and the rationale for the therapist's target 
behaviors. The therapist was provided with a handout of behavioral definitions (see 
Appendix D) in order to promote concurrent use and understanding of the language being 
used to define behaviors. 

The basic guidelines of a behavior analysis and therapy approach with an emphasis 
on differential reinforcement were then reviewed. Therapists were instructed to ignore all 
inappropriate behaviors, provide positive reinforcement in the form of praise and attention 
contingent upon compliance and positive behaviors, use time-out or a physical guide for 
noncompliance, use clear, concise and age appropriate commands to direct the child's 
behavior, and be consistent in the use of the techniques. This didactic and modeling phase 
typically took 90 minutes to complete. 
Phase two —Interactive training 

Following Phase 1 , the therapists practiced their skills by providing PT to non-target 
children. After a ten-minute warm-up period, the therapist's behavior was coded for the 
next five minutes without receiving prompts or feedback. The primary author provided 
prompting and feedback for the remainder of the session to help shape the therapist's 
appropriate use of the behavioral techniques. Additionally, target behaviors of the child 
were labeled to facilitate the identification of behavior problems. 

Following each PT session, the therapist received prompt verbal feedback on her 
overall performance. Criterion for discontinuing Phase 2 included the therapist providing 



37 

75% of all praise and attention contingent upon appropriate behavior, 75% of all requests 
for action were direct commands, and 75% of negative behaviors were ignored during the 
five minute segment that was coded. Therapist 2 had four practice sessions during this 
phase, and Therapists 1, 3, and 4 all had five practice sessions. Each practice session 
averaged one hour. 
Phase three -Feedback 

When the therapist reached criterion, she resumed PT with the target children from 
the baseline period. All sessions continued to be videotaped. Each session began with a 
brief reminder of the behavioral goals for that session based on the observations from the 
previous session. The therapist then received feedback only at the end of each session. 
Because of the time constraints on the therapists at All Children's Therapy Center, the 
therapists frequently received feedback from the author via telephone, or occasionally by fax. 
This phase lasted until the completion of the research. However two children were 
discharged prior to the end of this study and thus have limited data intervention. 
Follow Up 

A follow-up observation period was attempted with each therapist approximately 
one month after completing the training to test the hypothesis that the intervention allowed 
the therapists to generalize their skills to other non-target children. Each therapist was 
observed during 2-5 PT sessions with different children. Each session was videotaped and 
coded for therapist target behaviors only for comparison across baseline and intervention 
data. Because of difficulty in scheduling these sessions, and in recruiting appropriate 
participants, some follow-up sessions were not conducted until 8 months postintervention. 



RESULTS 

Percent occurrence or rate per minute was calculated for each of the child and 
therapist target behaviors. Percent occurrence was calculated by dividing the total frequency 
of a behavior by the total number of occurrences in the general behavior category for each 
session. For example, to calculate the percent occurrence of direct commands for each 
session, the frequency of direct commands was divided by the total number of commands 
(direct and indirect). Results are summarized first across all therapists and children in relation 
to the original hypotheses and then compared between the two settings in which the study 
was conducted. 

Therapist Behaviors 

Direct commands. The percent occurrence of direct commands for all dyads is 
shown in Figures 1 and 2. During the intervention phase, all four therapists exceeded the 
goal level for 75 percent of all commands to be direct. Prior to the intervention the average 
percent occurrence for direct commands was 62 percent (range = 42% - 80%) (see Table 3). 
During the intervention the average percent of direct commands increased to 83 percent 
(range = 79% - 87%). In the inpatient setting, therapists increased their direct commands an 
average of 27 percent, while therapists in the outpatient setting increased their direct 
commands by 15 percent. Therapist 1 (with Ron) and Therapist 4 (with Bryan) had already 
attained the goal level for direct commands prior to the intervention. Additionally, Therapist 
4 demonstrated an upward trend during baseline. As can be seen in the multiple baseline 



38 



39 



Table 3 



all therapists. 


Behavior Category 


Baseline 


Intervention 


Direct Commands 


62" o 


83% 


Compliance 


82% 


88% 


Noncompliance 


16% 


12% 


No opportunity for 
compliance 


55% 


39% 


Inappropriate child 
behavior 


1.4/min 


1/min 


Praise 


1/min 


2.7/min 


All attention to 
appropriate child behavior 


81% 


96% 


Inappropriate child 
behavior followed by 


20% 


78% 


ignore 







design, the increase in direct commands for Therapist 2 did not occur until after the 
intervention, and follows the increase in direct commands attained by Therapist 1 . The upward 
trend demonstrated by Therapist 4 with Bryan began after the intervention had already been 
implemented with Therapist 3. However, Therapist 4 did not reach the goal level with Mary 
during baseline and, in fact, had a downward trend with Mary. 

Inappropriate behavior followed by ignore. The percent occurrence for ignoring 
inappropriate child behaviors was calculated by dividing the number of inappropriate child 
behaviors that were ignored by the total number of inappropriate behaviors followed by ignore 
plus attention. The percent occurrence of ignoring inappropriate child behavior for each 
child/therapist dyad in the inpatient setting is shown in Figures 3 and 4. If there was no 
inappropriate behavior during a session to ignore, then no data point is present in the graph. All 
four therapists demonstrated significant increases in ignoring inappropriate behavior 



40 



Baseline 



Intervention 



100 n 

SO 

I 60 

1 40 -, 
20 - 




-i 1 1 r 



100 
80 
60 



I 40 



20 


100 

80 



| 60 
I 40 



Follow-up Probes 



Therapist 1/Sierra 



Non-target children 



-\ 1 1 1 1 1 1 1 1 I r 



Therapist 1/Sherry 



-e • •- 



Non-target children 



i 1 1 1 t 1 1 1 1 1 1 1 1 1 1 r 



20 




~[ 1 1 r 



n 1 1 1 1 1 1 



"l 1 1 1 1 1 



Therapist 1/Ron 



Non-target children 



~~\ 1 1 1 1 I 1 1 1 1 i r 



100 

80 



I 60 



20 



"i — i 1 1 1 — i 



Therapist 2/ Alan 



-e a *- 



Non-target children 



~i 1 1 1 1 r 



i 1 1 1 1 1 1 1 1 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 5 6 

Sessions 



Figure 1 : Percent occurrence of therapist direct commands-inpatient setting 



41 



Baseline 



Intervention 





iuu - 






80 - 




c 
u 
p 


60 - 

40 - 

20 - 

- 


i i 



t 1 r 



Therapist 3/Katy 



Non-target children 



~[ 1 1 r 



~i 1 r 



100 



80 



| 60 

ft, 40 



20 - 



n 1 1 r 



~i 1 1 r 



100 
80 



"£ 60 1 
I 40 



20 



Therapist 4/Bryan 



-e a »- 



Non-target children 



i I I i r 



~\ 1 r 



Therapist 4/Mary 
-• s +- 



Non-target children 



Follow-up Probes 



T 1 1 1 1 



• • ^ 



i f 1 1 1 1 1 1 1 1 1 r 



"i 1 1 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5 

Sessions 



Figure 2 : Percent occurrence of therapist direct commands— outpatient setting 



42 



Baseline 



Intervention 



Follow-up Probes 





100 




80 




u 


£ 


60 

40 




20 








~i 1 r 





100 




80 


c 
u 

Q- 


60 
40 




20 









100 




80 


C 

u 
p 

Eh 
U 

0- 


60 
40 




20 









i 1 - "~ i r 



-i 1 r 



Therapist 1/Sierra 



-® « ©- 



Non-target children 



-i 1 1 1 r 



Therapist 1 /Sherry 



Non-target children 



~\ 1 1 1 1 1 1 i i r 



Therapist 1/Ron 



Non-target children 



-r^— i 1 r 



H 1 1 1 1 r 



100 
80 



I 60 



20 




Therapist 2/Alan 



8 • G — 

Non-target children 



n 1 1 1 ; 1 1 1 1 



1 1 - * - ! 1 J~'—\ 1 1 1 1 1 1 1 — 

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 

Sessions 



Figure 3 : Percent occurrence of ignoring inappropriate child behavior— outpatient setting 



43 



Baseline 



Intervention 



Follow-up Probes 




Sessions 



Figure 4 : Percent occurrence of ignoring inappropriate child behavior- 
outpatient setting 



44 

during the intervention. During baseline, very little inappropriate behavior was ignored (M = 
20%; range = 9% - 30%). The goal for this category was to ignore 75 percent of all 
inappropriate child behavior. 

During the intervention, there was a dramatic increase by all therapists to an average 
rate of 78 percent (range = 65% - 91%). Therapists in the inpatient setting attained an average 
increase of 62 percent while the therapists in the outpatient setting attained an average increase 
of 53 percent. The average percent occurrence for this behavioral category at baseline and 
during the intervention for all therapists is shown in Table 3. Although all therapists 
demonstrated significant increases in ignoring inappropriate behaviors, the average percentage 
of two of the therapists from the outpatient setting did not meet the goal level with two of the 
children (Therapist 3 (T3)/Katy = 65%, & T4/Mary = 65%). Therapist 1 exhibited an obvious 
upward trend during baseline with Ron, which continued during the intervention and was stable 
during the last two sessions at 100 percent. Therapist 1 also exhibited a subtle upward trend 
with Sherry and Sierra. However, following the intervention, the mean shifts between baseline 
and intervention were more obvious for Sherry and Sierra. 

Attention to appropriate child behavior. Figures 5 and 6 show the total attention to 
appropriate child behavior, a combination of both attention and praise. The percent occurrence 
of attention to appropriate child behavior was calculated by dividing the total amount of 
attention to appropriate behavior (A+ and Praise) by the total amount of attention to both 
appropriate and inappropriate behaviors. During baseline, all therapists demonstrated high 
levels of attention to appropriate child behavior. Attention to appropriate child behavior was 
more variable during baseline, whereas during the intervention, the rates of attention were 
higher and more stable. 



45 



Baseline 



100 
80 A 



§ 6 ° 
I 40 



20 




100 

80 
| 60 
I 40 

20 




-i 1 1 r 



100 

80 
| 60 H 
I 40 

20 




i 1 1 r 



Intervention 
-• — .* — ♦ — —• -♦- 



Follow-up Probes 

i • « 



Therapist 1 /Sierra 



Non-target children 



i 1 1 1 1 1 1 1 1 1 1 1 h 1 i i i i i 



Therapist 1/Sherry 



Non-target children 



i 1 1 1 1 1 1 1 1 1 i i ( i i i i i i 



Therapist 1/Ron 



-e a •- 



Non-target children 



n 1 1 1 1 1 1 1 1 1 1 r 



100 

80 

1 60 

$ 40 

a. 

20 




n i t r 



~\ 1 1 1 1 1 



Therapist 2/Alan 



-e a o- 



Non-target children 



12 3 4 5 6 7 



1 1 — n 1 1 1 1 1 1 1 1 1 1 1 1 

9 10 11 12 13 14 15 16 17 1 2 3 4 5 6 
Sessions 



Figure 5 : Percent occurrence of therapist attention to appropriate child behavior— 
inpatient setting 



46 



Baseline 



Intervention 





100 




so 


c 


60 


u 




y 




— 




o 




CU 


40 



20 



-♦ — ,- 



Therapist 3/Katy 



Non-target children 





100 




SO 


— 

c 

p 

Pu 


60 

40 



20 



i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 r 1 1 1 i i 



100 

so 



c 60 
s 

I 40 A 



20 




— i — i 1 — i — i 1 — 

12 3 4 5 6 7 



■ ■ 



Therapist 4/Bryan 



Non-target children 



Follow-up Probes 



i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 i i i i 



Therapist 4/Mary 



• « f- 

Non-target children 



+ 



i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 

9 10 11 12 13 14 15 16 17 18 1 2 3 4 5 
Sessions 



Figure 6 : Percent occurrence of therapist attention to appropriate child behavior- 
outpatient setting 



47 

Therapists in the inpatient setting had an average increase of 17 percent during the 
intervention, while therapists in the outpatient setting had an average increase of 12 percent The 
mean occurrence of attention to appropriate behavior across all child/therapist dyads at 
baseline and intervention are provided in Table 3. Therapist 1 demonstrated an obvious upward 
trend during baseline with Ron that increased further during the intervention. However, 
Therapist 1 did not exhibit the same trend with Sherry or Sierra. Therapist 4 also demonstrated 
a slight increase in this target behavior during baseline with Mary that increased further 
during the intervention. 

Praise. Praise was defined as a separate and specific form of attention to appropriate 
child behavior. Praise was directed at child conduct not motor learning. The rates of praise per 
minute for each child/therapist dyad are depicted in Figures 7 and 8. As predicted, all four 
therapists demonstrated low levels of praise during baseline. Only after the implementation of 
the intervention did all therapists showed substantial improvements in providing praise. The 
mean rates of praises per minute for all therapists at preintervention and intervention are shown 
in Table 3. Before the intervention, therapists provided an average of 1 praise per minute 
(range = .42 - 2.1). During the intervention, there was an increase by all therapists to an average 
rate of 2.4 praises per minute (range = 1.8 - 3.8). The decline in the rate of praise for Therapist 
1 , with Ron, is likely the result of the increased independence Ron exhibited in completing his 
exercises prior to discharge. Ron's increasing independence with his treatment program 
resulted in decreased interactions with his therapist.. 

Child Behaviors 

Compliance. As a result of the therapists' training, it was hypothesized that child 
compliance rates would increase and inappropriate child behaviors would decrease. The percent 
occurrence of compliance across each therapist/child dyad is shown in Figures 9 and 10. 



48 



Baseline 



Intervention 



Follow-up Probes 



u 4 

I 3 
E 
5 2 

1 





Therapist 1/Sierra 



Non-target children 



l 1 1 1 1 1 1 1 1 1 1 r 



E 3 



$ 2 



u 
ed 



n i r 



Therapist 1/Sherry 



Non-target children 




n 1 1 1 r 



i 1 1 1 1 f 



3 - 




1 4- 

1 3- 

E 




| 2- 

3 1- 

- 


! 



~1 1 1 1 1 1 



Therapist 1/Ron 



Non-target children 



: 1 1 1 1 1 1 1 1 1 1 1 P 1 1 1 1 1 1 



5 i 
4 



J 3 
E 
S 2 

Qi 

1 





Therapist 2/ Alan 



Non-target children 



- 



~i 1 1 1 1 1 1 r 



n 1 1 1 1 r 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 5 6 

Sessions 



Figure 7 : Rate of therapist praise to child-inpatient setting 



49 



Baseline 



Intervention 



Follow-up Probes 



3 -1 



i 2 






-i 1 1 r 



Therapist 3/Katy 



Non-target children 




-I 1 1 1 1 1 T 1 1 I I 



4 - 



3 3 



I 2 



Therapist 4/Bryan 



Non-target children 



■ ■ 



n 1 r 



~i 1 1 1 1 






9 

K 

1 



Therapist 4/Mary 



-» s — — J- 



Non-target children 




~i 1 1 1 ; 1 1 1 r 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5 

Sessions 



Figure 8 : Rate of therapist praise to child— outpatient setting 



50 

The percent of commands complied with was calculated by dividing the number of 
commands complied with by the total number of commands followed by compliance plus 
noncompliance. Thus commands with no opportunity to comply were not included. There 
was variability for all of the children during baseline and the intervention, ranging as much as 
35 percent between sessions. Although the rates of compliance were relatively high to begin 
with, during the intervention, 5 of the 7 children improved their rates of compliance further. 
However, of the remaining two, one child (Sierra) had a small decrease in compliance and 
another child (Sherry) had no mean change overall. Two children (Mary & Sherry) had an 
upward trend during baseline that confounded the improvements seen during the 
intervention. The average percent occurrence of compliance for both baseline and 
intervention across all children is shown in Table 3. 

Noncompliance. Figures 11 and 12 show the percent occurrence for child 
noncompliance with therapist commands. Rates of noncompliance were calculated by 
dividing the frequency of noncompliance by the total number of commands given with 
opportunities to comply. During the intervention, five of the children (Ron, Alan, Katy, 
Bryan and Mary), exhibited declines in their rates of noncompliance. Although one of the 
graphs (Sherry) indicated a downward trend in noncompliance, the two spikes in 
noncompliance during sessions 6 and 9 resulted in no change in the mean rate of 
noncompliance overall. Another child's rate of noncompliance (Sierra) declined throughout 
the intervention, however her mean rate of noncompliance across all sessions increased 
from a baseline rate of 7 percent to 1 5 percent during the intervention. 

Inappropriate child behaviors. Figures 13 and 14 show the rates of inappropriate 
child behavior per minute across all therapist/child dyads. There was tremendous variability 
in the frequency of inappropriate child behaviors across sessions during both baseline and 



51 



Baseline 



Intervention 



Follow-up Probes 



100 -I 




80 - 


\ 


1 6 °" 
1 40- 




20 - 
- 


1 I 1 



c 
u 
p 

— 

u 

CL 



100 

80 
60 - 
40 - 
20 




-i 1 1 1 1 1 1 1 i i r 



n 1 1 r 



100 
80 H 

60 




Therapist 1/Sierra 



Non-target children 



n 1 1 1 1 1 



1 1 1 r 



-i 1 1 1 1 r 



B 
U 

y 



I 40 H 

20 




-i 1 1 r 



i 1 1 1 1 1 i i i i <~ 



B 
o 
u 

'_ 
o 

a. 



100 
80 
60 - 
40 - 
20 




Therapist 1 /Sherry 



Non-target children 



~\ 1 1 1 1 



>: o s 



Therapist 1/Ron 



Non-target children 



-\ 1 1 1 1 



Therapist 2/Alan 



-» a •- 



Non-target children 



— i 1 1 1 1 1 

12 3 4 5 6 7 



~i 1 r 



9 10 11 12 13 14 15 16 17 1 2 3 4 5 6 
Sessions 



Figure 9 : Percent occurrence of child compliance— inpatient setting 



52 



100 



80 



Baseline 



Intervention 



-4— 1 

a 
u 

a 

a; 

a, 


60 
40 




20 








Follow-up Probes 



-[ 1 1 — 

3 5 



n 1 1 1 1 

7 9 11 



-i 1 1 1 1 r 

13 15 17 



100 



80 



* 60 

o 
Q 

i- 
6 

* 40 



20 



t 1 1 1 1 1 1 r 





100 




80 


— 


60 


a; 




U 




c 




1) 




Oh 


40 



20 



Therapist 3/Katy 



Non-target children 



1 1 1 1 1 

19 21 23 



n 1 1 1 1 1 1 r 



Therapist 4/Bryan 



Non-target children 



• + 



Therapist 4/Mary 
-e s — — ♦- 



Non-target children 



n 1 r 



n 1 1 1 1 1 1 1 1 1 1 1 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5 

Sessions 

Figure 10 : Percent occurrence of child compliance— outpatient setting 



53 



Baseline 



Intervention 



70 
60 
50 

| 40 

fc 30 -\ 

On 

20 H 
10 




t 1 1 1 r 



~i 1 1 1 r* - ! - * - ] 1 1 1 r 



Follow-up Probes 



Therapist 1/Sierra 



Non-target children 



~i 1 1 1 




1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 5 6 

Sessions 



Figure 1 1 : Percent occurrence of child noncompliance— inpatient setting 



54 



Baseline 



Intervention 



70 
60 
50 



§ 4(H 
J 30 H 



20 
10 H 




Follow-up Probes 




70 
60 
50 



§ 40 

o 



20 

10 



70 
60 
50 



1 40 
p 

£ 30 H 



20 

10 





n 1 r 



Therapist 3/Katy 



Non-target children 



~i 1 1 1 I i I I r 




~i 1 1 1 1 r 




Therapist 4/Bryan 



Non-target children 



i 1 1 r~ — i r 



Therapist 4/Mary 



Non-target children 



n 1 1 1 



~i 1 1 1 1 



1 1 1 1 1 1 r 

12 3 4 5 6 7 



I I I T 



1 I I 



9 10 11 12 13 14 15 16 17 18 1 2 3 4 5 
Sessions 



Figure 12 : Percent occurrence of child noncompliance— outpatient setting 



55 



the intervention for six of the seven children. On average, the frequency of inappropriate 
behaviors across all children during the intervention decreased from 1.4 per minute (range = 1 - 
1.7) to 1 per minute (range = .2 - 1.7) (see Table 3). However, for two children (Katy, and Mary), 
the average rate of inappropriate behaviors during the intervention increased. The average 
decrease in inappropriate behavior was greater than the average increase. Specifically, for the two 
children whose rates of inappropriate behavior increased, the average increase in frequency was .2 
inappropriate behaviors per minute (range = .1 - .3) while the average decrease in the frequency 
of inappropriate behaviors per session for the four other children was .8 (range = .5 - 1.1). For 
one of the children (Alan), there was a very large and stable reduction in the occurrence of 
inappropriate behavior. Katy, on the other hand, demonstrated an increase in inappropriate 
behavior. The increases in her inappropriate behavior appear to be an extinction burst directly 
related to increases in the frequency of ignoring inappropriate behaviors by her therapist. Mary 
also demonstrated a tendency towards increasing inappropriate behaviors; however, the limited 
data points for Mary during the intervention preclude making strong conclusions about her 
behavior. Sierra's inappropriate behavior showed an increase during baseline with an upward 
trend during the intervention, although the mean rate of inappropriate behaviors did not change. 
Ron demonstrated a downward trend during baseline that continued during the intervention; 
however, his awareness of his impending discharge may have contributed to the decrease in his 
inappropriate behavior. 

Follow-up. Follow-up was conducted to determine if the behavior management skills 
the therapists learned in the intervention would generalize to non-target children. Follow-up data 
for therapist target behavior are shown in Figures 1-8. Mean occurrence of target behavior is 
found in Table 3. Follow-up data for therapist target behaviors indicate that the majority of 



56 



Baseline 



Intervention 



Follow-up Probes 




1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5 

Sessions 



Figure 13 : Rate of inappropriate child behavior—inpatient setting 



57 



4i 
3.5 
3 

1 2.5 -\ 

c 

I 2 

I 1-5 

1 

0.5 




4 

3.5 

3 

tj 2.5 

c 

I 2 

at lJ 

1 

0.5 




4 i 
3.5 
3 H 
£ 2.5 



2 - 






£ 1.5 



1 

0.5 





Baseline 



Intervention 



n 1 1 r 



Therapist 3/Katy 





n 1 1 1 1 1 1 r 




Therapist 4/Bryan 



Non-target children 



■ — • 



~i 1 r 



Therapist 4/Mary 



Non-target children 




Follow-up Probes 



-i 1 1 1 1 1 1 1 1 i r 



i 1 1 1 i 



n i r 



12 3 4 5 6 7 



~i 1 1 1 1 1 1 1 1 1 1 1 1 1 i 

9 10 11 12 13 14 15 16 17 18 1 2 3 4 5 

Sessions 



Figure 14 : Rate of inappropriate child behavior— outpatient setting 



58 

the skills learned from the intervention were generalized to non-targeted children 
postintervention. Specifically, direct commands, praise and attention to appropriate 
behaviors remained at or above the criterion level. During the follow-up sessions, the levels 
of praise decreased somewhat for most therapists but still remained well above the baseline 
level. The frequency of praises per minute decreased during follow-up for Therapist 4 and 
Therapist 2 but increased further for Therapist 1 and Therapist 3. All therapists' follow-up 
levels of praise remained well above their baseline levels. The percent occurrence for direct 
commands fell during follow-up sessions for Therapist 1 when compared with the rates of 
direct commands with all three target children. However, the average rate of direct 
commands remained well above Therapist 1 's baseline level achieved with Sherry and Sierra. 
The rate of direct commands also fell during follow-up sessions with Therapist 2, but again 
the follow-up level remained well above the average achieved at baseline. 

Attention to appropriate child behavior remained high and stable for all therapists. 
The ability to ignore inappropriate child behaviors fluctuated among the therapists. 
Although the average rate of ignoring remained higher than the baseline levels, it fell for all 
therapists. This may have been due to the very low rate of inappropriate behaviors exhibited 
by the follow-up participants (1-2 inappropriate behaviors per session). 

Comparisons between settings. Table 4 shows the average change on several 
behavior categories for therapists and children in both settings. Overall, the therapists made 
positive changes in the target behaviors in both settings. As a result of the therapists' 
changes, the majority of child behaviors also improved. However, the degree of change of 
behavior varied in each setting. Therapists in the outpatient setting had lower rates of 
ignoring inappropriate child behavior compared to therapists in the inpatient setting. On 
other target behaviors, they attained very similar rates of behavior although they may have 



59 



Table 4 

Average change in the inpatient setting vs. outpatient setting . 



Behavior Category 



Average change - 
inpatient setting 



Average change - 
outpatient setting 



Direct commands 


27% increase 


15% increase 


Child compliance 


3% increase 


11% increase 


Child noncompliance 





10% decrease 


Commands with no 






opportunity for 


17% decrease 


15% decrease 


compliance 






Praise 


1.5/minute increase 


1.9/minute increase 


Attention to appropriate 
child behavior 


.3/minute decrease 


.4/per minute increase 


Inappropriate child 
behavior 


.7/minute decrease 


(i 


Inappropriate child 






behavior followed bv 


62% increase 


52% increase 


ignore 







started at different levels at baseline. For example, therapists in the inpatient setting made a 
27 percent increase in their direct commands across all children; however, this group of 
therapists started with a rate of direct commands 14 percent lower than that of the 
outpatient therapists. Overall, both groups attained a similar average rate of direct 
commands (see Table 5). 

Children in the outpatient setting had a larger increase in compliance (11%) 
compared to the children in the inpatient setting (3%). However, children in the outpatient 
setting started with a 10 percent lower rate of compliance in the inpatient setting. Yet both 
groups reached a similar level of compliance during the intervention phase. Additionally, 



60 



Table 5. Average rate of occurrence for target behavior at pre- and post-intervention and 
across settings 



Inpatient Setting 



Outpatient Setting 



Behavior Category 


Pre- 
intervention 


Post- 
intervention 


Pre- 
intervention 


Post- 
intervention 


Direct command 


56% 


83% 


70% 


84% 


Compliance 


86% 


89% 


76% 


87% 


Noncompliance 


11% 


11% 


24% 


14% 


No opportunity for 


55% 


38% 


55% 


40% 


compliance 










Praise 


1.0/min 


2.5/min 


.98/min 


2.9/min 


Inappropriate child 
behavior 


1.4/min 


.7/min 


1.5/min 


1.5/min 


All attention to 


81% 


98% 


82% 


94% 


appropriate child 
behavior 










Inappropriate child 
behavior followed by 


23% 


85% 


16% 


68% 


ignore 











children in the outpatient setting had a larger decrease in noncompliance (10%), whereas 
setting experienced no change, on average, in rates of noncompliance. However, once again, 
children in the outpatient setting had higher rates of noncompliance during baseline (24%) 
than children in the inpatient setting (11%). 

Consumer Satisfaction 
On the measure of consumer satisfaction (see Appendix B), it was predicted that the 
therapists receiving SIT would rate the training as acceptable, effective and beneficial in 
reducing behaviors problems and increasing compliance in children during physical therapy. 
Table 6 shows the mean scores on the satisfaction questionnaire. Overall, the items on the 
questionnaire were responded to positively except the time commitment needed for the 
study was on average rated as "a little long". Additionally, one therapist rated the behavior 



61 

of one child as "somewhat worse" (Therapist 1 /Sierra). Two therapists commented that the 
most helpful aspects of the training included "learning to reinforce positive behavior and to 
ignore negative behavior", another therapist reported that "definite guidelines to follow 
when there was a compliance problem" was most helpful. When asked to describe the least 
helpful aspects of the training, one therapist reported "nothing really". Another therapist 
reported that "time out just did not fit in this setting, nor did time out or a physical guide 
work with all kids or all ages." Two therapists made no suggestions for improvement. 

Table 6 

Mean scores on Consumer Satisfaction Questionnaire . 

Item Mean Range 

1. Rationale for the training 

2. Relevance of training to work 

3. Learning techniques for dealing with behavior 

4. Time commitment needed for the training 

5. Acceptability of techniques 

6. Child's behavior following training 

7. Child's compliance following training 

8. Changes observed in non-target children 

9. Intervention facilitated more productive PT 

10. Overall rating of training (10-point scale) 



4.75 


(4-5) 


4.25 


(4-5) 


4.25 


(4-5) 


2.00 


(1-3) 


4.25 


(3-5) 


4.14 


(2-5) 


4.43 


(4-5) 


4.50 


(4-5) 


4.50 


(4-5) 


8.25 


(7-9) 



DISCUSSION 

This study sought to determine if physical therapists could learn general behavior 
analysis and therapy skills using an intervention program that incorporated a modeling 
videotape, individual interaction training, feedback and reinforcement. An additional 
purpose of this study was to determine if the physical therapist's use of behavior analysis and 
therapy would result in decreased levels of inappropriate child behaviors and increased levels 
of compliance during PT. A multiple baseline design was used with two therapists and four 
children in an inpatient children's hospital, and two therapists and three children in an 
outpatient PT setting. 

Results indicated that after participating in Strategic Interaction Training (SIT), the 
physical therapists learned the behavior analysis and therapy skills and incorporated them 
into their daily PT sessions and, that the changes in their behavior were the result of the 
intervention. Specifically, following the intervention, therapists increased their levels of 
direct commands, ignoring inappropriate child behavior, attention to appropriate child 
behavior, and praise. As a result, compliance increased, and noncompliance and 
inappropriate behaviors decreased for most children. Results also indicated differences in the 
amount of change in target behaviors between participants in the inpatient setting and the 
outpatient setting. The intervention was rated positively by the therapists. 

Ignoring inappropriate child behavior seemed to be the most difficult behavior 
change for therapists to make. Two therapists did not reach the goal level during the 
intervention, and the rates for this behavior fell during follow-up for all therapists. One 



62 



63 

possibility for this finding is that the goal level was too high. However, it is worth noting 
that the average percent occurrence of inappropriate child behaviors increased for the two 
children whose therapists did not reach the goal level set for ignoring inappropriate child 
behaviors. It seems likely that the intermittent reinforcement in the form of therapist 
attention to the inappropriate behaviors contributed to the increase in inappropriate child 
behaviors. The children whose therapist's average percent of change in ignoring 
inappropriate behavior was highest, and thus had a more consistent rate of ignoring, had the 
greatest decrease in inappropriate behaviors. Katy appeared to experience an extinction 
burst for her inappropriate behaviors. During sessions when the rate of therapist ignoring 
was high, the rate of child inappropriate behaviors was also high. When the rate of therapist 
ignoring was low, the rates of inappropriate child behavior were also low. Thus, it may be 
that for this dyad, the therapist was punished with an increase in inappropriate child 
behaviors when she ignored more consistently. 

During the intervention, therapists were coached and given feedback to provide a 
high level of target behaviors. Following the intervention, the rates of some therapist target 
behaviors, such as direct commands, ignoring inappropriate child behavior, and praise, fell 
for most therapists. Follow-up sessions were conducted at least one month post- 
intervention. During that time, it is likely that therapists assumed a more natural rate 
of target behaviors. Whereas a more consistent rate of ignoring would be best for 
extinguishing inappropriate child behaviors, an intermittent rate of positive reinforcement 
in the form of praise would have a better effect on the maintenance of appropriate 
child behaviors. 



64 



Five of the seven children demonstrated an increase in compliance to therapist 
commands following the intervention. Only one child had a decrease in compliance and one 
child's rate of compliance stayed the same. It is difficult to determine cause and effect for 
changes in a particular behavior when an intervention targets multiple behaviors, and several 
factors may have influenced the positive changes in compliance rates. First, all therapists 
made increases in the rates of direct commands. Direct commands tell the child specifically 
and clearly what to do versus indirect commands that may inadvertently present a command 
as if the child had a choice about complying. Second, the increase in compliance to 
commands may be the result of a decrease in rates of commands given with no opportunity 
to comply. It seems probable that eliminating multiple commands for the same behavior, 
and allowing the child time to comply, would result in the increased compliance rates. 

On the consumer satisfaction questionnaire or in comments made to the 
experimenter, all therapists commented on the use of time out being difficult to utilize due 
to time constraints or the inability to enforce time out through the use of a restraint process. 
Although time out has been used extensively and effectively with children with 
developmental disabilities and behavior problems (Olmi, Sevier, & Nastasi, 1997), the use of 
time out may not have been appropriate for these settings because of the issues of time and 
restraint. Handen, Parrish, McClung, Kerwin, and Evans (1 992) found time out to be more 
effective in promoting compliance in children with mild developmental delays when 
compared to a physical guide. A physical guide is often perceived as less aversive, easier to 
administer, and does not negatively reinforce the child by allowing the child to escape from a 
difficult situation (Handen et al., 1992). The physical guide was viewed differently by 
different therapists in this study. One therapist, who only used physical guides and not time 



65 

out, commented that she liked having definite guidelines to follow when there was a 
compliance problem. Another therapist reported that the physical guide did not seem to 
work with older children. 

Differences Between Settings 

Differences between the settings included the outpatient setting being a more 
structured setting where therapists had sequential appointments throughout the day with 
very little flexibility in their schedules. This had implications for the amount of feedback 
they received as well as the format in which they received their feedback. Because the 
therapist usually had another child waiting for the next appointment, these therapists 
frequently did not have time immediately following a session with a target child to receive 
and discuss the feedback. As a result, the researcher had to provide feedback in different 
formats such as via telephone and written feedback via fax. Therapists in the inpatient 
setting had a great deal more flexibility in their schedules and were available to receive and 
discuss verbal feedback following each session with the target children. 

Another aspect of the research that was different between the settings was the length 
of the relationship each therapist/child dyad had prior to their involvement in the study. 
Children in the outpatient setting had been working with their therapists, for over 1-2 years. 
Their behavioral repertoires were strongly engrained and perhaps more resistant to change. 
These lengthy relationships may account for the higher rates of noncompliance and lower 
rates of compliance seen during baseline. In the inpatient setting, the children had been 
working with their therapists for only 1-2 months. These relationships were still forming and 
the therapists may have been more lenient, using more indirect commands versus direct 
commands, and the children may have been more eager to please, possibly contributing to 
the higher rates of compliance seen during baseline. 



66 

Another aspect of the research that was different between settings was the presence 
of a research assistant or the primary investigator. During the inpatient intervention, the 
primary investigator (PI) provided all coaching and feedback during the training and 
intervention, with a research assistant observing. Due to significant time constraints, the PI 
was not always able to provide the coaching and feedback directly to the outpatient 
therapists. Instead, the PI frequently informed the research assistant as to what behaviors to 
coach a therapist in during Phase 2 of the intervention and what feedback to give during 
Phase 3. The PI did conduct Phase 1 of the intervention with all therapists. 

Additionally, the environment the children were in when not participating in PT and 
the frequency of PT with the target children differed between the two PT settings. In the 
inpatient setting, children were patients in the hospital for many weeks, away from their 
families, schools, and homes. The hospital was the environment in which they were 
residing. Additionally, these children were hospitalized for rehabilitative purposes and they 
received PT every day. On the other hand, children being seen in the outpatient setting were 
living in their own homes with their families, going to their own school every day and 
coming to the PT setting for their therapy. These children had different needs for therapy 
and were seen once, twice, or at most, three times a week. These children could also miss 
their therapy and frequently did. These variations between the settings may have impacted 
the rates of change in various behaviors. In spite of the fact that there were several key 
differences between the settings, all therapists involved in the study demonstrated positive 
changes on target behaviors. This lends support for the generalizability of the intervention 
across different settings and therapy formats. 



67 



Strengths and Weaknesses of the Study 

Strengths of this study include conducting the study in multiple settings. 
Additionally, the variety of children, with varying diagnoses and behavior problems indicate 
the generalizability of this intervention in promoting positive changes in both children and 
therapist's behaviors. Incorporating follow-up sessions that occurred at least one month 
postintervention with primarily different, non-target children, further demonstrated the 
generalization of the intervention as well as the maintenance of the skills learned. 

An additional strength of the study is the involvement of several therapists. Few 
studies have involved therapists in the intervention, and then it has only been one therapist 
that has been involved. This study is the first study to train multiple therapists in behavior 
analysis and therapy. 

Weaknesses of the study include differences in the implementation of various 
components of the study. The greater use of the research assistant in the outpatient setting 
may have accounted for some of the differences in the outcome. However, this also speaks 
to the likelihood that such training could be conducted by less well-trained personnel, on a 
less frequent basis, perhaps resulting in a more cost-efficient intervention. 

Additionally, although this study had more therapists involved in the intervention 
that any previous study, 4 therapists and 7 children is still a small sample size. A further 
weakness is that two children left the study prematurely during the intervention. In the 
inpatient setting, Ron was told he would be going home just prior to implementing the 
intervention with his therapist. This information may have had a positive impact on his 
behavior above and beyond that of the intervention. In the outpatient setting, Mary left 
for a six-week vacation with her family and thus we were only able to obtain three 
intervention data points. 



68 

The use of a multiple baseline design has both strengths and weaknesses. For the 
purposes of this study, a new intervention, it allows for a detailed analysis of both PT and 
child behavior. It also allows for ongoing assessment of the intervention as well as 
refinement and change in the intervention depending on the data. However, while there are 
certain guidelines used with graphic analysis, the reader's interpretations of the findings may 
be influenced by their own experiences and views regarding the procedures, meaning, and 
importance of the research (Parsonson & Baer, 1992). Additionally, there are concerns 
regarding external validity, and the generalizability of the findings outside the sample. 
Replication of the results in future studies can increase the confidence of these findings. 

Implications 

This study is an important contribution to both fields of physical therapy and 
psychology. It describes an effective training intervention that psychologists can use to help 
physical therapists learn behavior analysis and therapy skills to enhance productive physical 
therapy. It is the first study to train multiple physical therapists on the use of general 
behavior management skills that can be applied with many children with a variety of 
behavior problems. Such techniques will allow the therapist to work more autonomously 
and deal more effectively with children who present with difficult behavior. 

Few studies have incorporated training the physical therapists to use a behavioral 
intervention in the PT setting. The majority of studies have typically involved a behavior 
specialist to conduct the intervention, leaving the therapist dependent on a behavior 
specialist each time they encounter a difficult child whose behaviors interfere with PT. This 
study extends the current literature by demonstrating that physical therapists can learn 
behavior management skills and effectively implement these skills into their PT sessions. 



69 

Additionally, this study demonstrated the effectiveness of the intervention in two very 
diverse PT settings, inpatient and outpatient, which demonstrates the generalizability 
of the findings. 

This study has further implications for the future of collaborative efforts between 
physical therapy and psychology. Overt, disruptive behavior in children is but one aspect in 
which psychologists could help facilitate improved physical therapy. Other problem 
behaviors can consist of depression, anxiety, fear, chronic pain, noncompliance, coping with 
loss (of limb or function), and death and dying issues. Through better integration of mind 
and body functioning through increased collaborations between PTs and psychologists, it is 
likely that PTs can further enhance the effectiveness of their interventions with both 
children and adults facing a myriad of health problems. 

Future Directions 

This study clearly demonstrates that Strategic Interaction Training resulted in 
positive changes in the therapist's behaviors, and positive changes in the majority of target 
child behaviors. However, noncompliance and inappropriate child behaviors that were 
disruptive to the treatment process still occurred during PT. Having the ability to identify 
specific variables maintaining noncompliance and problem behaviors would likely enhance a 
therapist's ability to further impact the occurrence of these problems. Recently, researchers 
have been training other professionals such as teachers, and parents to conduct their own 
functional analyses in the classroom or at home and to generate hypotheses of the functional 
relationships of the aberrant behaviors being exhibited (Foster-Johnson & Dunlap, 1 993; 
Frea, Koegel, & Koegel, 1993). 

Boggs, Danforth, & Stokes (1986) demonstrated that mothers of children with 
attention deficit hyperactivity disorder could learn to provide contingent rewarding, clear 



70 

instructions, and positive attention to their child and that as a result, child compliance rates 
increased and inappropriate child behaviors decreased. However, untreated child behaviors 
were not affected until the parents were trained to analyze functionally their child's behavior. 

Therapists in this study were not trained in a manner that would promote 
generalization (Stokes & Baer, 1977). Having learned basic behavioral techniques during SIT, 
the ability to analyze functionally child problem behavior would have allowed therapists to 
apply more readily the principles and techniques learned during SIT to new behaviors 
exhibited by other non-target children during therapy. Such training would be more 
consistent with generalization programming promoted by Stokes and Osnes, 1989. The 
better equipped the therapist is to assess the functional contingencies of behavior, and utilize 
that information, the more likely that treatment techniques will be effective. 

This seems a logical next step in training other professionals in behavior analysis and 
therapy. In this study, although positive changes were evident for the majority of target 
behaviors, individualized approaches for some of the children would likely have resulted in 
greater behavior changes. For example, Sherry had both legs immobilized for six weeks in 
long leg casts following her surgery. During her admission for rehabilitation, she 
demonstrated extreme anxiety about bending her knees. Her inappropriate behaviors, 
including aggression, noncompliance and screaming, escalated during activities that focused 
on bending her knees. An individualized approach could incorporate relaxation training 
exercises such as deep breathing, imagery, and positive self-talk, outside of the PT setting to 
provide her with better coping skills during PT. However, not only have physical therapists 
not received training in behavioral techniques to deal with problem behavior, they also rarely 
receive training in integrating cognitive-behavioral techniques such as relaxation training into 
their therapy. 



71 

Had the therapist been able to conduct a functional analysis for Sierra, it likely would 
have demonstrated that many of Sierra's most difficult behavior problems escalated when 
she was asked to perform a particular exercise that was more painful than the others. In this 
situation, if the therapist followed Sierra's noncompliance with a time out, her behavior was 
reinforced with escape from the difficult and painful exercise. Again, an individualized 
approach could have incorporated additional techniques such as relaxation training and 
teaching Sierra different ways to perform the exercise independently, thus providing her 
more self-control during difficult exercises. 

Physical therapists frequently conduct exercises that are painful for their patients. 
Many patients may experience fear or anxiety during PT because of the discomfort they 
experience. Physical therapist would benefit from knowing how to utilize cognitive- 
behavioral techniques such as relaxation training, deep breathing, and imagery. The 
integration of psychological interventions with physical interventions can only serve to 
further enhance the effectiveness of therapy, decrease some of the aversiveness of therapy, 
and empower patients by encouraging them to take more control over their pain, anxiety and 
fears. Health and pediatric psychologists could be instrumental in promoting this training 
and providing ongoing consultation for such issues. 

Parent training is the most commonly implemented behavior management program. 
It has been well-established that a group format can be equally if not more effective than 
individual training (Eyberg & Matarazzo, 1980; Webster-Stratton, 1990). The format used 
for Strategic Interaction Training could easily be adapted for use in a group format for future 
use. This format would have the advantages of sharing problems and intervention ideas 
among therapists and be a more efficient use of both the psychologist's and therapist's time. 
Additionally, child behavior problems are certainly not limited to the PT setting. 



72 

Occupational therapists, speech therapist, nurses, and doctors, as well as the child's parents, 
could all potentially benefit from similar training in behavior management techniques. 
Training multiple exemplars outside of the PT setting would allow for generalization of 
child's improved behavior outside of the PT setting as well (Stokes & Baer, 1977). Because 
consistency is so important in effecting change in children's behavior, it seems likely that if a 
child's behavior is treated consistently by a variety of people he/she comes into contact with, 
it would result in further reductions in inappropriate and noncompliance behaviors. 

Finally, now that it has been established that physical therapists can learn behavior 
therapy skills and effectively implement these skills, it will be meaningful to assess if PT was 
in fact more productive in terms of attaining PT goals in a more timely manner following a 
training intervention in behavior analysis and therapy. The benefit of such an intervention 
to decrease the number of PT sessions needed for some children and increase the 
effectiveness of PT should be assessed in future studies. Additionally, the cost effectiveness 
and cost offset of such an intervention needs to be evaluated. For children who do not 
receive PT due to severe behavior problems, or whose PT does not progress due to their 
problem behaviors, a behavioral intervention such as SIT would be clearly indicated, and yet 
is infrequently employed due to lack of knowledge by the therapist. 

A future study to measure progress made towards PT goals is warranted. Such a 
study could take the form of an experimental design involving two groups of children 
receiving PT, who are matched with similar disabilities and PT goals, and exhibit significant 
disruptive behaviors that interfere with PT. The experimental group would receive PT from 
therapists who have received SIT, while the control group would continue to receive 
standard PT from untrained therapists. The number of sessions to reach the therapeutic 
goals would be the dependent variable. 



73 

Several single-subject designs have been conducted with children who were not 
progressing in therapy until a behavioral intervention was implemented (Horner, 1971; 
Chandler & Adams, 1972). These studies lend support to the efficacy of a behavioral 
intervention for children with behavior problems that impede progress in physical therapy. 
However, more controlled studies, with larger samples may need to be conducted to further 
promote the benefits of incorporating behavior analysis and therapy into the PT setting. 



APPENDIX A 
PROCEDURAL OUTLINE - SIT TEACHING SESSION 

Before Session: 

1 . Set up room; 

a) Put modeling videotape in VCR and set up television. 

b) Have handouts on Behavioral Definitions, Key points about Ignoring, and time 
out. 

c) Have video camera set up with tape and microphone turned on. 
Treatment Session 

1. Provide overview of SIT. 

PHASE 1: Today we will talk about certain behaviors and behavioral definitions 
and watch a videotape of a physical therapist working with a child who has some 
behavior problems. 

PHASE 2: Starting tomorrow, your PT sessions will be observed, coded, and 
videotaped with non-participating children. I will be coding different intervals in 
each session. I will also be helping to label target behaviors of the child, and provide 
prompting for you to use target the behaviors we discuss today. At the end of each 
session, we will review the data on your performance. Your goal is to make 75% of 
all praise and attention contingent upon appropriate behavior, ignore at least 75% of 
all negative behavior, and to make at least 75% of all requests for action direct 
commands. 

PHASE 3: When the goals as stated above are met, PT will resume therapy with 
children from the baseline phase. Each session will be observed, coded, and 
videotaped. After each session, you will receive feedback on the target behaviors. 

2 Provide handout of behavioral definitions to therapist. Review each definition 

and the examples provided before reviewing the tape. Attention will focus on the 
therapist's target behaviors noted from baseline. 

3. Commands should be direct rather than indirect. 

a) A direct command is a clearly stated order, demand, or direction in declarative 
form. 

b) The statement must be sufficiently specific as to indicate the behavior that is 
expected from the child 



74 



75 



c) Give examples of "let's", "how about", and "why don't you" commands. 

Examples: "Lets do your leg lifts now." "Can you walk now?" "Lets transfer 
to the mat, OK?" "How about ten more?" 

• Many indirect commands are expressed in question form. 

• A direct command should leave no question in the child's mind that he/she 
is being told to do something, no illusion of choice. 

• You can still be polite and preface the command with "please". Give 
examples of direct commands: "Transfer to the mat please." "Please take 
five steps forward." 

4 Commands should be positively stated. 

a) Tell the child what "to do" instead of what "not to do". 

b) Try to avoid "don't" commands. In many situations, it is possible to give a 
positively stated command that is incompatible with the negative behavior you 
are trying to eliminate (e.g., "Straighten your arm" instead of "Don't bend your 
elbow." Or "Put the toy down" instead of "Don't throw that." 

5 Commands should be simple. 

a) Commands should be things that the child is capable of doing as much as 

possible. However, I realize that is difficult to always know what a child in PT is 
physically capable of doing, and that as a PT, you are frequently working to 
increase the child's ability. 

6. Commands should be given one at a time. 

a) Children have a hard time remembering more than one thing at a time. Avoid 
stringing together commands (e.g., "Go transfer onto that mat, do 10 straight- 
leg-raises, 10 heel slides, and 10 short-arc-quads." That's a lot to ask a child to 
do all at once. 

b) Instead break that big command down into its smaller parts. 

c) Another problem in giving a big command is that the child has an awful lot of 
work to do before he/she gets positive feedback from you for complying. We 
will talk about compliance and praise in detail in a few minutes. 

7 Commands should not be vague. 

a) Make sure that your commands tell the child specifically what to do. 

b) Commands like "be careful", "relax", "be good", "look" are so nonspecific that 
the child does not know exactly what to do in order to comply. Also, commands 
such as "Do your bridges" does not tell the child how many to do. Instead use 
commands like "Put both crutches on the floor please." "Walk to the door 
please", "Do ten bridges". 

8. Indirect Commands - are orders, demands, or directions for a behavioral response 

that is implied, nonspecific, or stated in question form. This type of command 
implies that the child has a choice when often they do not. Provide examples. "Lets 



76 



walk now", "You might want to sit down now", "How about we do your leg lifts 
first?", "Can you reach for the bean bag?" 

9. Ask the therapist what he/she would do if the child complied with the 

therapists commands. 



a) Instruct them to give the child a labeled praise contingent upon child 
compliance. 

b) A praise is any specific or nonspecific verbalization that indicates liking, 
approval, or expresses a favorable judgment upon an activity, product, or 
attribute of the child. 

c) Distinguish b/w labeled and unlabeled praise. 

d) Give examples of praise in terms of compliance: 

• "You did a great job sitting up all by yourself (labeled praise) 

• "I like how well you are walking today" (labeled praise) 

• "Thank you for transferring to the mat so quickly!" (labeled praise) 

• "Good job." "Way to go!" (unlabeled praise) 

10. Ask the therapist to provide examples of appropriate behaviors that occur in 

PT. List the examples provided by the physical therapist: 



a) Discuss how the therapist's attention (verbal, as in talking, or physical touch) 
should only be delivered as long as the child is behaving appropriately (e.g., as 
long as the child is doing his/her exercises, or the child is calm while being 
stretched, the PT can provide attention to him/her). 

b) Emphasize how this is different from praise. 

c) Encourage therapist not to talk following a command, until the command is 
complied with. 

d) Discuss how this might interfere with learning new motor patterns, as well as just 
being distracting and interrupting, which may interfere with the child completing 
the command. 

11 Ask the therapist to provide examples of negative behavior that occur in PT. 

List the examples the PT provides: 



77 



a) Reinterpret these as compliance, noncompliance, inappropriate behaviors, and 
escape techniques. 

b) List your respowse/reinterpretation next to the example of negative behavior 
the PT provides. 

c) The therapist will probably list many behaviors that are attention seeking 
behaviors or escape motivated behaviors. Identify these. 

d) Inform therapist that all inappropriate behaviors are to be IGNORED. 

e) Provide Ignoring handout. 

a) When you ignore negative behavior you give no verbal or nonverbal reaction, and 

continue what you are doing, except for providing further talk other than to give 
commands. 

b) At first, when you ignore negative behavior, it may get worse. 

c) But with CONSISTENCY on your part the inappropriate behavior will go 
away. 

d) Be sure to be consistent in your ignoring or it will not work. Give example of 
child wanting candy in the grocery store. If you do not think you can 
consistendy ignore a behavior, do not begin to ignore, or stop ignoring early 
rather than later. 

e) As soon as the negative behavior ceases, provide attendon and praise. 
13 Discuss Compliance and Noncompliance. 

a) If you give a command and the child is noncompliant, meaning he/she does not 
initiate the command in 5 seconds and complete the behavior without further 
prompting, you are to physically guide the child through the command or give 
the warning discussed below. 

b) As long as the child is continuing to follow through with the command, 
regardless of inappropriate accompanying behavior such as crying, even after 5 
seconds, it is compliance. 

14 Discuss Time out vs. Physical Guide. 

a) If the child is noncompliant after five seconds, you may physically guide the child 
through the motion, or command, showing no emotion and without comment. 
Then, immediately provide another simple direct command and praise the child 
if he/she complies or pause at least five seconds before providing attention to 
the child, and then provide attention only to appropriate behavior. 

b) If the therapist opts for time out, instruct the therapist to give the following 

warning if the child disobeys: "Ifjou don 't .you 're going to sit in time out. " 

Do not repeat your command or ask if he/she heard you. You need to use 
these exact words each time. Again, after the warning, you need to decide if the 
child is being noncompliant. 

15 Ask the therapist what he/she would do if the child obeyed the warning. 

Instruct the therapist to give a labeled praise. 



78 



16 Instruct the therapist in the time out procedure using the time out diagram. 

a) Place the child on the mat and say only 'You didn't do what I told you to do so you 
have to stay here on the mat. " Ask the therapist what he/she thinks the child will do 
at this point. 

b) Discuss issues about time out, safety, mobility of the child, etc. 

c) Discuss how to get the child to the mat quickly and safely, possibly carrying the 
child from behind. 

d) Once the child is on the mat, the therapist is to say "Now stay here until I tell you to 
get off." The child has to stay in the room for exacdy three minutes with five 
seconds of silence before the therapist can return. 

17. Discuss the importance of using those exact words and nothing extra. The 
words were chosen to be the shortest simplest way to tell the child the reason he/she 

is in time out and what he/she has to do now. The child gets no extra attention. 
Time out is effective only if the therapist is in control of when the child can leave 
time out. 

18. When the child is in time out, what kinds of things could he/she do? Discuss 

these issues and how to handle them. Basically, anything the child does is 
ignored unless the child's safety is at risk. 

19 Define the following: 

a) Inappropriate Behaviors - includes crying, whining, or yelling, and smart talk. 
Crying consists of inarticulate utterances of distress (audible weeping) at or 
below the loudness of normal conversation. Whining consists of words uttered 
by the child in a slurring, nasal, high-pitched, falsetto voice. Yelling consists of a 
loud screech, scream, shout or loud crying. The sound must be loud enough so 
that it is clearly above the intensity of normal indoor conversation. Smart talk 
consists of is impudent or disrespectful speech. 

b) Aggression - verbalization of threats, even if said in a joking tone of voice, and 
physical threats or behavior directed at (1) therapist, (2) self, or (3) environment. 

c) Pain behaviors - grimacing, vocalized complaints of pain, such as yelling out, 
whining about pain. 

20 Discuss Attention to Inappropriate Behavior. 

a) Attention to inappropriate behavior includes talking to the child, reacting by 
pulling away, or stopping an activity, providing physical touch, other than is 
therapeutically necessary, when the child is being noncompliant, crying/ whining, 
dawdling, or being aggressive. This also includes negotiating, "threatening" the 
child, or allowing the child to escape from a demand situation, activity, or 
therapy. 



79 



Examples: 

• Child is purposefully falling in the walker, and therapist says "Stand up!" 

• Child is having a tantrum and the therapist tries to calm him/her by talking 
or touching, or allows the child a break. As soon as the child calms down, 
and complies, he/she can then have a break. 

• Child is refusing to scoot back on the mat, and therapist continues to say 
"Scoot back!" 

b.)Describe how attention to inappropriate behaviors increases the likelihood of 
increasing inappropriate behavior. 

21. Describe Information Descriptions and Behavioral Descriptions. 

a) Information descriptions introduce information about people, objects, events, or 
activities, but do not clearly describe the child's current or immediately 
completed behavior. 

Examples of Information Descriptions: 

"I can tell you have been doing your exercises". 

"Tomorrow is Saturday." 

"I'm going to work on your legs first." 

"Your legs feel tight today". 

"This is the last set". 

b) Behavioral Descriptions are statements used to describe the child and a verb 
describes the child's ongoing or immediately completed verbal or nonverbal 
observable behavior. 

Examples of Behavioral Descriptions: 

• "You're transferring to the mat." 

• "I see you're trying hard to lift that leg." 

• 'You're pushing your wheelchair." 

22 Review videotape. 

a) Point out examples of praise, direct commands, ignoring negative or 
inappropriate behaviors, and attention to appropriate behaviors. 

23 Answer any questions. Write down what questions are asked. 



80 



24. Review basic rules of behavior modification. 

a) Ignore all negative behaviors. 

b) Review "Key Points About Ignoring" handout. 

c) Provide positive reinforcement contingent upon positive behavior. 

d) Use direct commands to increase compliance with requests for action. 

e) Use a lot of praise following compliance for commands. 

f) Consistency is key!!! 



APPENDIX B 



DEMOGRAPHICS QUESTIONNAIRE 



Participants Initials: 

Age: Gender: 

Diagnoses: 



Brief medical history: 



Subject #_ 



Date: 



Goals for physical therapy: 



Physical Therapists initials: 

Number of years practicing PT: 

Number of years working with children:. 
Number of years at Shriner's: 



Highest degree obtained: 

NDT Training? Yes No 



81 



APPENDIX C 
CONSUMER SATISFACTION QUESTIONNAIRE 

Please circle the response for each question that best expresses how you 
honestly feel. 

1. Regarding the goals/rationale for the training, I feel that they were: 



1. not stated 2. poorly stated 3. stated 

somewhat 



4. stated well 5. stated very 

clearly 



2. Regarding the relevance of the training to my work, I feel that the goals for the training 
were: 



1. not at all 2. a little relevant 3. somewhat 

relevant relevant 



4. very relevant 5. extremely 
relevant 



3. Regarding the techniques for dealing with negative behavior of children in PT, I feel I 
have learned: 



1. nothing 2. very litde 3. a few 

techniques techniques 



4. several useful 5. many useful 
techniques techniques 



4. Regarding the time commitment needed for the training, I feel it was: 
1. too long 2. a litde long 3. just right 4. short 

5. How acceptable are the techniques you learned in the training to you? 

1. not at all 2. somewhat 3. neutral 4. acceptable 

acceptable acceptable 

Please describe: 



5. very short 
(would like 
more) 



5. very 

acceptable 



6. After the training, how would you describe the patient's behavior 

1. considerably 2. somewhat worse 3. the same 4. somewhat 5. gready 

worse improved improved 



82 



83 



7. After the training, I felt that my patient's compliance to my commands or requests was: 

1. considerably 2. somewhat worse 3. the same 4. somewhat 5. gready 

worse improved improved 

8. How would you describe any behavior changes in the children you work with as a result of 
the techniques you learned? 

1. much worse 2. somewhat worse 3. the same 4. somewhat 5. gready 

than before than before improved improved 

9. How much do you feel that the techniques you learned helped facilitate more productive 
physical therapy with the children you worked with? 

1. not at all 2. verylitde 3. no change 4. somewhat 5. very 

helpful helpful 

1 0. Overall, how would you rate the training you have received to deal with negative behavior 
in physical therapy? 

12345 67 89 10 

not at all helpful helpful extremely 

helpful 

1 1 . What aspect of the training do you feel was most helpful? 



12. What aspect of the training was least helpful? 



13. Please provide any feedback you feel would be helpful enhance the Strategic Interaction 
Training program 



APPENDIX D 
BEHAVIORAL DEFINITIONS 

Direct command - is a clearly stated order, demand, or direction in declarative form. The 
statement must be sufficiently specific as to indicate the behavior that is expected from the 
child. Commands should be direct vs. indirect, positively stated, simple, concise, and given 
one at a time. 

Examples: 

• "Lift your leg up." 

• "Transfer to the mat." 

• "Do ten more leg lifts please." 

• "Take five steps forward please." 

• "Walk to the door please." 

Indirect command - is an order, demand, or direction for a behavioral response that is 
implied, nonspecific, or stated in question form. This type of command implies that the 
child has a choice when often they do not. 
Examples: 

• Lets do your leg lifts now. 

• Do you want to walk now? 

• Lets transfer to the mat, OK? 

• How about ten more? 

Compliance - Compliance with instructions or commands. Child begins to comply 
following a direct command, or an indirect command within 5 seconds after it is given and 
completes the behavior without further prompting. Any negative behavior occurring while 
the child is complying does not affect compliance and should be ignored. Ex: PT says "Lift 
your leg up" and child says "NO!" but lifts leg anyway. 

Noncompliance - Child does not begin to complete an instruction, or starts but fails to 
complete the command, direct or indirect, within ten seconds of the request. This includes 
delay tactics that interfere with therapy - behaviors that allow a delay in activity. 



84 



85 



Examples: 

• self-stimulating behaviors (rocking, touching self, etc.) 

• engaging others in interaction, such as saying "hi" to people in the room/hall 

• arguing with therapist, parents, or others 

• distractibility, not paying attention, 

• commenting/ conversing about events or objects in environment 

• making an excuse, or just saying "NO". 

• any diversion from the task at hand that interferes with the child initiating the 
command within ten seconds and completing the command without further 
prompting. 

No Opportunity for Compliance - After a command is given, the child is given 5 seconds 
to respond. No opportunity for compliance occurs when the child has not been given an 
adequate chance to comply. 
Examples 

• Commands that request a behavior to be performed in the not-immediate future 
(> 5 sec). 

♦"You can ride the bike when we finish." 

♦Child dumps out all the bean bags and PT says "You have to put them away 
when you 
are done." 

• When therapist completes the action requested in the therapists command, thus 
preventing the child from complying. Ex: PT says "Lift up your leg." And then 
picks up the child's leg for him/her. 

• Verb phrases in a command that do not provide enough information for the 
child to perform the expected behavior. 

• "Look" "Listen" 

• "Be careful" can be replaced with a direct command "Put both crutches 
on the floor." 

• "Wait a minute." Can be replaced with a direct command "Stop at the 
door. 

Praise - any specific or nonspecific verbalization that indicates liking, approval, or expresses 
a favorable judgment upon an activity, product, or attribute of the child. 
Examples: 

• You did a great job sitting up all by yourself! (labeled praise) 

• I like how well you are walking today, (labeled praise) 

• Thank you for transferring to the mat so quickly! (labeled praise) 

• Good job. (unlabeled praise) 

• Wow! (unlabeled praise) 

• Way to go! (unlabeled praise) 

Attention following a ppropriate child behavior - Talk or physical contact with the child 
contingent upon appropriate child behavior. 



86 



Examples: 

• Child is ambulating well, and therapist talks to him/her. 

• Child completes a set of exercises well, and therapists pats his/her arm 
and praises his/her efforts 

• Child is whining, therapist ignores (see below), child stops crying, and 
then therapist provides attention after child stops crying. 

• Child is tolerating stretching and is not whining, and therapist uses 
distraction by telling child a story. 

• If the child is whining but complying, the therapist can continue to 
interact and praise the child for the positive behavior that is occurring. 
Ex: PT says "Walk to the door." Child complies with the command, but 
whines, therapist can praise the compliance and ignore the whining. 

Inappropriate Behavior - includes crying, whining, or yelling, and smart talk. Crying 
consists of inarticulate utterances of distress (audible weeping) at or below the loudness of 
normal conversation. Whining consists of words uttered by the child in a slurring, nasal, 
high-pitched, falsetto voice. Yelling consists of a loud screech, scream, shout or loud crying. 
The sound must be loud enough so that it is clearly above the intensity of normal indoor 
conversation. Smart talk consists of is impudent or disrespectful speech. 
Examples: 

Do I haaavwweeee to? NO! (very loud; yelling) 

This is too hard. STOP IT! (very loud; yelling) 

I don't want to do this anymore. Your stupid, (smart talk) 

Why should I? (smart talk) You can't make me! (smart talk) 

I don't like you anymore! (smart talk) Inappropriate grunting/noise 
I'm gonna get you back (smart talk) Stop or I'll hit you! 

Aggression - any physical touch that is intended to be antagonistic, aversive, hurtful, or 
restrictive of the therapists activity or aggressive behavior directed at (2) self, or (3) 
environment. 

Examples: 

• Ripping up a book 

• Breaking or throwing a toy 

• hitting or attempting to hit someone or something (pounding fist in mat), swats 
at therapist or parent 

• Refusing walker and kicking it over or grabbing the walker out of PT's hand. 

• child pinches therapist 

• child pushes therapists hand away 



Pain behaviors - grimacing, vocalized complaints of pain, such as yelling out. 
Examples: 

"OWWWW!" 
"That hurts!!" 
"That's far enough!!!! That's far enough!!!" 



8^ 



Attention following inappropriate child behavior - Talking to the child, looking at the 
child or providing physical touch, other than is therapeutically necessary, when the child is 
exhibiting inappropriate behavior, is noncompliant, or is being aggressive. This would 
include negotiating, such as offering a reward such as a desired toy, the presence of a parent, 
or an activity contingent on performance, "threatening" (a specific verbalization indicating a 
negative consequence will follow a behavior) the child, or allowing the child to escape from 
a demand situation, activity, or therapy. 
Examples: 

• Child is yelling, and therapist continues to talk and try to calm him/her. 

• Child has stopped ambulating to ask a question, and therapist answers. 

• Child is screaming inappropriately, and therapist is providing comfort. 

• Child is being noncompliant and therapist says "If you walk to the door 
then we will stop." (negotiating) 

• Child is not standing appropriately and therapist says "You can play this 
game only if you stand up." (negotiating) 

• If you don't stop crying mommy will have to leave, (threat) 

• If you throw that toy you will never play with it again, (threat) 

• Child starts whining during an exercise and therapist responds "When 
you stop whining, we will continue." 

• Child is noncompliant with therapists requests and is sent back to room, 
(escape) 

Ignore inappropriate behavior. There is active ignoring and passive ignoring. During active 
ignoring, the therapist may take action by withdrawing physically from the child, restrain the 
child or remove an object from the child. During passive ignoring, the therapist does not 
provide any evidence of having heard or seen the inappropriate behavior that occurs, and 
works through the behavior. Often times, the child is allowed to escape from an activity by 
using aggression such as trying to prevent you from stretching by pushing your hands away. 
Always try to resume the activity, even if for a short duration, so that the child is not 
reinforced for aggression or other negative behavior. 

Examples of Passive Ignoring: 

• Child whines, and therapist does not react in any way, or acknowledge the child's 
whining. Does not ask child to stop crying or ask what questions. Therapist 
continues with the exercise. 

• Child acts out by purposefully not doing SLR's correctly; therapist may not pay 
any attention to this, does not attempt to correct. 

• Child throws toy; therapist does not scold or acknowledge that behavior 
happened; leaves the toy on the floor (passive). 

• Child tries to prevent therapist from bending his/her knee further by pushing 
therapists hand away and PT continues to bend knee, not 

acknowledging/ addressing child's behavior. When child stops resisting for ~ 2 
seconds, therapist can stop bending the knee. This way, the child is not 
reinforced by getting the PT to stop when he/she wants. 



88 



• Child stops walking and begins crying. Therapist can physically guide the child 
and push the walker forward briskly to complete the task. The crying is not an 
issue and is not addressed. When the task is completed, the therapist can still 
acknowledge completion: 

• PT - "Walk to the door." 

• child is walking to the door, and is almost there when he/she begins to 
cry and stops walking 

• PT briskly and firmly guides the child to the door with no emotion, 
comments, or extraneous physical touch. 

• Once at the door, PT can say "You walked to the door" Does not 
comment on crying. 

• Child is complying with a command but is crying or screaming. PT ignores the 
crying and praises the child's compliance. 

• PT: "Do ten bridges." 

• child: "No!" but begins doing them 

• PT: "One, good job doing your bridges. Two. . ." 

Examples of Active Ignoring: 

• Child acts out by purposefully not doing SLR's correctly; PT physically holds the 
child's leg preventing the exercise, but does not address the child. Once child 
stops the negative behavior, physical therapist can provide a direct command 
such as "Tighten this muscle first, keep your knee straight and lift your leg. OR, 
therapist can firmly guide the child through the SLR's, but not address the child's 
behavior. 

• Child acts like he/she will throw a toy, and therapist removes the toy from the 
child's hand and move it away, without speaking or other acknowledgment 
(active). 

Information Descriptions - introduce information about people, objects, events, or 
activities, but do not clearly describe the child's current or immediately completed behavior. 



Examples of Information Descriptions: 

"I can tell you have been doing your exercises". 

"Tomorrow is Saturday." 

"I'm going to work on your legs first." 

"Your legs feel tight today". 

"This is the last set". 

Behavioral Descriptions - are statements used to describe the child and a verb describes 
the child's ongoing or immediately completed verbal or nonverbal observable behavior. 
Examples of Behavioral Descriptions: 

• 'You're transferring to the mat." 

• "I see you're trying hard to lift that leg." 

• 'You're pushing your wheelchair." 



REFERENCES 

Allen, K. D., & Stokes, T. F. (1987). Use of escape and reward in the management of 
young children during dental treatment. Journal of Applied Behavior Analysis . 20 . 381-390. 

Babbitt, R. L., Hoch, T. A., Coe, D. A, Cataldo, M. F., Kelly, K. J., Stackhouse, C, & 
Perman, J. A. (1994). Behavioral assessment and treatment of pediatric feeding disorders. 
Developmental and Behavioral Pediatrics. 15 . 278-291. 

Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current dimensions of applied 
behavior analysis. Journal of Applied Behavior Analysis . 20 . 313-328. 

Bandura, A. (1969). Principles of behavior modification . New York: Holt, Rinehart and 
Winston. 

Bandura, A. (1971). Psychotherapy based on modeling principles. In A. E. Bergin & S. 
L. Garfield (Eds.), Handbook of psychotherapy and behavior change (pp. 653-708). New 
York: Wiley. 

Bandura, A., & Kupers, C.J. (1964). Transmission of patterns of self-reinforcement 
through modeling. Journal of Abnormal and Social Psychology . 69 . 1-9. 

Baron, R. A. (1970). Attraction toward the model and model's competence as 
determinants of adult imitative behavior. Journal of Personality and Social Psychology . 14 . 
345-351. 

Bernstein, G. S. (1982). Training behavior change agents: A conceptual review. 
Behavior Therapy . 13 . 1-23. 

Bessmer, J. L. (1996). The dyadic parent-child interaction coding system II (DPICS-II): 
reliability and validity (Doctoral dissertation, University of Florida, 1996). Dissertation 
Abstracts International . 58-07B . p. 3961. 

Boggs, S. R., Danforth, J. S., & Stokes, T. (1986 ). Functional problem-solving skills: 
Increasing the generality of parent training . Paper presented at the meeting of the American 
Psychological Association, Washington, DC. 



89 



90 



Bragg, J. H., Houser, C, & Schumaker, J. (1975). Behavior modification: Effects of 
reverse tailor sitting in children with cerebral palsy. Physical Therapy . 55 . 860-868. 

Branch,, W. T. (1990). Teaching models in an ambulatory training program. Journal of 
General Internal Medicine . 5, SI 5-26. 

Briener, J., & Beck, S. (1984). Parents as change agents in the management of their 
developmentally delayed children's noncompliant behaviors: A critical review. Applied 
Research in Mental Retardation . 5, 259-278. 

Budd, K. S., Green, D. R., & Baer, D. M. (1976). An analysis of multiple misplaced 
parental social contingencies. Journal of Applied Behavior Analysis . 9, 459-470. 

Carr, E. G. (1977). The motivation of self-injurious behavior: A review of some 
hypotheses. Psychological Bulletin . 84, 800-816. 

Carr, E. G., & Durand, V. M. (1985). Reducing behavior problems through functional 
communication training. Journal of Applied Behavior Analysis . 18 . 111-126. 

Carr, E. G., Newsom, C. D., & Binkoff, J. A. (1980). Escape as a factor in the 
aggressive behavior of two retarded children. Journal of Applied Behavior Analysis , 
13,101-117. 

Carroll, W. R., & Bandura, A. (1985). Role of timing of visual monitoring and motor 
rehearsal in observing learning of action patterns. Journal of Motor Behavior . 
17,269-281. 

Chandler, L. S., & Adams, M. A. (1972). Multiply handicapped child motivated for 
ambulation through behavior modification. Physical Therapy . 52 . 399-401. 

Charlop, M. H., & Milstein, J. P. (1989). Teaching autistic children conversational 
speech using video modeling. Journal of Applied Behavior Analysis . 22, 275-285. 

Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and 
Psychological Measurement . 20 . 37-46. 

Cunningham, C. E., Davis, J. R., Bremner, R., Dunn, K. W., & Rzasa, T. (1993). Coping 
modeling problem solving versus mastery modeling: effects on adherence, in-session 
process, and skill acquisition in a residential parent-training program. Journal of Consulting 
and Clinical Psychology . 61, 871-877. 

Day, H. M, Horner, R. H, & O'Neill, E. O. (1994). Multiple functions of problem 
behaviors: Assessment and intervention. Journal of Applied Behavior Analysis . 27, 279-289. 



91 



Delamater, A. M., Conners, C. K., & Wells, K. C. (1984). A comparison of staff training 
procedures. Behavioral applications in the child psychiatric inpatient setting. Behavior 
Modification . 8, 39-58. 

Dowrick, P. W. (1991). Practical guide to using video in the behavioral sciences . New 
York: Wiley Interscience. 

Durand, V. M., & Carr, E. G. (1985). Self-injurious behavior: motivating conditions 
and guidelines for treatment. School Psychology Review . 14. 171-176. 

Eisenstadt, T. H., Eyberg, S. M., McNeil, C. B., Newcomb, K., & Funderburk, B. 
(1993). Parent-child interaction therapy with behavior problem children: Relative 
effectiveness of two stages and overall treatment outcome. Journal of Clinical and Child 
Psychology . 22. 42-51. 

Eyberg, S., Bessmer, J., Newcomb, K., Edwards, D., & Robinson, E. (1994). Dyadic 
Parent-Child Interaction Coding System-H: A Manual . Unpublished Manuscript, University 
of Florida, Gainesville, FL. 

Eyberg, S. M., Boggs, S. R., & Algina, J. (1995). New developments in psychosocial, 
pharmacological, and combined treatments of conduct disorders in aggressive children. 
Psychopharmacology Bulletin . 31 . 83-91. 

Eyberg, S., Edwards, D., Bessmer, J., & Litwins, N. (1994). The workbook: A coder 
training manual for the Dyadic Parent-Child Interaction Coding System II. Unpublished 
manuscript, University of Florida, Gainesville, FL. 

Eyberg, S. M., & Matarazzo, R. G. (1980). Training parents as therapists: a comparison 
between individual parent-child interaction training and parent group didactic training. 
Journal of Clinical Psychology . 36 . 492-499. 

Eyberg, S. M., & Robinson, E. A. (1982). Parent-child interaction training: Effects on 
family functioning. Journal of Clinical and Child Psychology . 11 . 130-137. 

Eyberg, S. M., & Robinson, E. A. (1983). Dyadic Parent-Child Interaction Coding 
System: A manual. Psychological Documents . 13 . Ms. #2582. 

Fabry, P. L., & Reid, D. H. (1978). Teaching foster grandparents to train severely 
handicapped persons. Journal of Applied Behavior Analysis . 11 . 111-123. 

Fordyce, W. E., Shelton, J. L., & Dundore, D. E. (1982). The modification of avoidance 
learning pain behaviors. Journal of Behavioral Medicine . 5, 450-414. 

Forehand, R. L., & King, H. E. (1977). Noncompkant children: Effects of parent 
training on behavior and attitude change. Behavior Modification . 1, 93-108. 



92 



Foss, B. M. (1966). Operant conditioning in the control of movements. Developmental 
Medicine and Child Neurology . 8, 339-340. 

Foster-Johnson, L., & Dunlap, G. (1993). Using functional assessment to develop 
effective, individualized interventions for challenging behaviors. Teaching Exceptional 
Children . 25, 44-50. 

Frea, W. D., Koegel, L. K., & Koegel, R. L. (1994). Understanding why problem 
behaviors occur: A guide for assisting parents in assessing causes of behavior and designing 
treatment plans . Santa Barbara: University of California. 

Fuller, P. R. (1949). Operant conditioning of a vegetative human organism. American 
Journal of Psychology . 62 . 587-591. 

Gardner, J. M. (1972). Teaching behavior modification to nonprofessionals. Journal of 
Applied Behavior Analysis . 5, 517-521. 

Handen, B. L., Parrish, J. M., McClung, T. J., Kerwin, M. E., & Evans, L. (1992). Using 
guided compliance versus time out to promote child compliance: a preliminary comparative 
analysis in an analogue context. Research in Developmental Disabilities . 13, 157-170. 

Harris, F. R.Johnston, M. K., Kelley, C. S., & Wolf, M. M. (1964). Effects of positive 
social reinforcement of regressed crawling of a nursery school child. Journal of Educational 
Psychology . 55, 35-41. 

Hendrickson, J. M., Gardner, N., Kaiser, P. & Riley, A. (1993). Evaluation of a social 
interaction coaching program in an integrated day-care setting. Journal of Applied Behavior 
Analysis . 26.213-225. 

Hester, S. B. (1981). Effects of behavioral modification on the standing and walking 
deficiencies of a profoundly retarded child. Physical Therapy . 61, 907-91 1. 

Hill, L. D. (1985). Contributions of behavior modification to cerebral palsy habilitation. 
Physical Therapy . 65, 341-345. 

Horner, R. D., (1971). Establishing the use of crutches by a mentally retarded spina 
bifida child. Journal of Applied Behavior Analysis . 4, 183-189. 

Hosford, R. E., & Johnson, M. E. (1983). A comparison of self-observation, self- 
modeling, and practice without video feedback for improving counselor interviewing 
behaviors. Counselor Education and Supervision . 23 . 62-70. 

Hudson, A. M. (1982). Training parents of developmental handicapped children: A 
component analysis. Behavior Therapy . 13 . 325-333. 



93 



Isaacs, C. D., Embry, L. H., & Baer, D. M. (1982). Training family therapists: an 
experimental analysis. Journal of Applied Behavior Analysis . 15 . 505-520. 

Iwata, B. A., Pace, G. M., Dorsey, M. F., Zarcone, J. R., Vollmer, T. R, Smith, R. G., 
Rodgers, T. A., Lerman, D. C, Shore, B. A., Mazaleski, J. L., Goh, H., Cowdery, G. E., 
Kalsher, M. J., McCosh, K. C, & Willis, K. D. (1994). The functions of self-injurious 
behavior: An experimental-epidemiological analysis. Journal of Applied Behavior Analysis . 
27,215-240. 

Kazdin, A. E. (1974). The effect of model identity and fear-relevant similarity on covert 
modeling. Behavior Therapy . 1, 427-452. 

Kazdin, A. E. (1994). Behavior modification in applied settings . Belmont, CA: 
Brooks/Cole. 

Koldene, M. L. (1971). Behavior Modification in the treatment of children with cerebral 
palsy. PT, 51, 1083-1091. 

Lalli, J. S., Mauk, J. E., Goh, H., & Merlino, J. (1994). Successful behavioral 
intervention to treat children who are reluctant to ambulate. Developmental Medicine and 
Child Neurology . 36, 625-629. 

Mace, F. C, Page, T. J., Ivancic, M. T, & O'Brien, S. (1986). Analysis of environmental 
determinants of aggression and disruption in mentally retarded children. Applied Research in 
Mental Retardation . 7, 203-221. 

Manella, K. J., & Varni, J. W. (1981). Behavioral therapy in a gait-training program for a 
child with myelomeningocele. Physical Therapy . 61. 1284-1287. 

Martin, J. A. (1976). Behavior modification and cerebral palsy. Journal of Pediatric 
Psychology . 1 , 48-50. 

Matson, J. L., & Stephens, R. M. (1978). Increasing appropriate behavior of explosive 
chronic psychiatric patients with a social-skills training package. Behavior Modification . 2, 
61-76. 

McCullagh, P. (1986). Model status as a determinant of observational learning and 
performance. Journal of Sport Psychology . 8, 319-331. 

McMahon, R. J., & Forehand, R. L. (1983). Consumer satisfaction in behavioral 
treatment of children: Types, issues, and recommendations. Behavior Therapy . 14, 209-225. 

Meichenbaum, D. H. (1971). Examination of model characteristics in reducing 
avoidance behavior. Journal of Personality and Social Psychology . 17, 298-307. 



94 



Meyerson, L., Kerr, N., & Michael, J. L. (1967). Behavior modification in rehabilitation. 
In S. W. Bijou & D. M. Baer (Eds.), Child development: Readings in experimental analysis 
(pp. 214-239). New York: Meredith Publishing Company. 

Miltenberger, R. G., & Veltum, L. G. (1988). Evaluation of an instructions and 
modeling procedure for training behavioral assessment interviewing. Journal of Behavioral 
and Experimental Psychiatry . 19 . 31-41. 

O'Dell, S. L. (1974). Training parents in behavior modification: a review. Psychological 
Bulletin . 81. 418-433. 

O'Dell, S. L. (1985). Progress in parent training. In M. Hersen, R. M. Eisler, & P. M. 
Miller (Eds.), Progress in Behavior Modification (vol. 19, pp. 57-108). Academic Press. 

Olmi, D. J., Sevier, R. C, & Nastasi, D. F. (1997). Time -in/time-out as a response to 
noncompliance and inappropriate behavior with children with disabilities: Two case studies. 
Psychology in the School . 34 . 31-39. 

Osnes, P. G. & Stokes, T. F. (1987, December). SCOOP: A direct observation 
instrument for the assessment of interactions . Presented at the Rivendell 
Conference for Clinical Practitioners, Memphis, Tennessee. 

Parsonson, B. S., & Baer, D. M. (1986). The graphic analysis of data. In A. Poling, & R. 
W. Fuqua, (Eds.), Research methods in applied behavior analysis (157-186). New York: 
Plenum Press. 

Parsonson, B. S., & Baer, D. M. (1992). The visual analysis of data, and current research 
into the stimuli controlling it. In T. R. Kratochwill, & J. R. Levin (Eds.), Single-case research 
design and analysis (pp. 15-40). Hillsdale, NJ: Lawrence Erlbaum Associates. 

Pierce, C. H., & Garland, D. (1977). Effects of four training procedures on increasing 
motor skills of physically handicapped retarded persons. Journal of Mental Deficiency 
Research . 21 . 25-35. 

Rapport, M. D., & Bailey, J. S. (1985). Behavioral PT and spina bifida: a case study. 
Journal of Pediatric Psychology . 10 . 87-97. 

Repp, A. C, & Karsh, K. G. (1994). Hypothesis-based interventions for tantrum 
behaviors of persons with developmental disabilities in school settings. Journal of Behavior 
Analysis . 27. 21-31. 

Reynell, J. K. (1965). Post-operative disturbances observed in children with cerebral 
palsy. Developmental Medicine and Child Neurology . 7, 360-376. 



95 



Rice, H. K., McDaniel, M. W., & Denney, S. L. (1962). Operant conditioning 
techniques for use in the physical rehabilitation of the multiply handicapped retarded patient. 
Physical Therapy . 48, 342-346. 

Riley, A. W., Parrish, J. M., & Cataldo, M. F. (1989). Training parents to meet the needs 
of children with medical or physical handicaps. In C. E. Schafer, & J. M. Briesmeister, (Eds.): 
Handbook of parent training: Parents as co-therapists for children's behavior problems (pp. 
305-336). New York: Wiley. 

Singer, L. T., Nofer, J. A., Benson-Szekely, L. J., & Brooks, L. J. (1991). Behavioral 
assessment and management of food refusal in children with cystic fibrosis. Journal of 
Developmental and Behavioral Pediatrics . 12 . 115-120. 

Stokes, T. F., & Baer, D. M. (1977). An implicit technology of generalization. Journal of 
Applied Behavior Analysis . 10 . 349-367. 

Stokes, T. F, & Kennedy, S. H. (1980). Reducing child uncooperative behavior during 
dental treatment through modeling and reinforcement. Journal of Applied Behavior 
Analysis . 13. 41-49. 

Stokes, T. & Mowery, D. (1999, May) OTIS observation system . Tampa: de la Parte 
Florida Mental Health Institute. (In revision). 

Stokes, T., Mowery, D., Dean, K., & Hoffman, S. J. (1997). Nurturance traps of 
aggression, depression, and regression affecting childhood illness. In D. M. Baer & E. M. 
Pinkerston. (Eds.). Environmental approaches to social problems (pp. 147-154). Boulder, 
CO: Westview Press. 

Stokes, T. F., & Osnes, P. G. (1988). The developing applied technology of 
generalization and maintenance. In R. H. Horner, & G. Dunlap (Eds.). (1988). 
Generalization and maintenance: Life-style changes in applied settings, (pp. 5-19). Baltimore, 
MD, USA: Paul H. Brookes Publishing. 

Thelen, M. H, Fry, R. A., Fehrenbach, P. A, & Frautschi, N. M. (1979). Therapeutic 
videotape and film modeling: A review. Psychological Bulletin . 86 . 701-720. 

Trotter, A. B., & Inman, D. A. (1968). The use of positive reinforcement in PT. 
Physical Therapy . 48, 347-352. 

Webster-Stratton, C. (1981). Videotape modeling: A method of parent education. 
Journal of Clinical Child Psychology . 1_0, 93-98. 

Webster-Stratton, C. (1990). Enhancing the effectiveness of self-administered videotape 
parent training for families with conduct-problem children. Journal of Abnormal Child 
Psychology . 18, 479-492. 



BIOGRAPHICAL SKETCH 
StaceyJ. Hoffman is the daughter of William R. Hoffman and Sandra S. Hoffman, 
Ph.D. She was born on November 6, 1965, in Elmhurst, Illinois. Stacey graduated from Paul 
D. Schreiber High School in January of 1983. She received a Bachelor of Science degree in 
physical therapy in 1987, at the Medical College of Georgia in Augusta, Georgia. 

Stacey worked at Duke University Medical Center in Durham, North Carolina, for 
four years as a pediatric physical therapist. While in Durham, she discovered a passion for 
Whitewater kayaking, and after a summer at the Nantahala Outdoor Center in Wesser, North 
Carolina, she moved there in 1991. She continued to paddle while working as a pediatric 
physical therapist at the Developmental Evaluation Center in Cullowhee, North Carolina, 
and began taking psychology courses. 

In 1993, Stacey began graduate studies in the Department of Clinical and Health 
Psychology at the University of Florida specializing in pediatric psychology. She completed 
her predoctoral internship at the Louis de la Parte Florida Mental Health Institute at the 
University of South Florida. As part of her internship, she conducted policy research in the 
area of mental health needs for children with special health care needs. Stacey currendy 
resides in Melrose, Florida while working as a postdoctoral fellow in pediatric psychology at 
the University of Florida. 



96 



I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and 
quality, as a dissertation for the degree of Doctor of Philosophy. 




Stephen R. Boggs, Chair 
Associate Professor of Clinical 
and Health Psychology 

I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and 
quality, as a dissertation for the degree of Doctor of Philosophy. 

Sheila Eyberg I 

Professor of Clinical and Health 
Psychology 

I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and 
quality, as a dissertation for the degree of Doctor of Philosophy. 




KathyeTight ~^~ 
Associate Professor of Physical 
Therapy 

I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and 
quality, as a dissertation for the degree of Doctor of Philosophy. 




Mary Jane Rapport ' 
Assistant Research Professor of 
Special Education 



I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and 
quality, as a dissertation for the degree of Doctor of Philosophy. 




Trevor Stokes 

Professor of Child and Family 

Studies 
University of South Florida 



This dissertation was submitted to the Graduate Faculty of the College of 
Health Professions and to the Graduate School and was accepted as partial fulfillment 
of the requirements for the degree of Doctor of Philosophy. 



December, 1999 .-^C^*^^: fd^^— C^ /? «,„ ?/*««*) 

Dean, College of Health 
Professions 



Dean, Graduate School 



UNIVERSITY OF FLORIDA 



3 1262 08555 2999